id	sid	tid	token	lemma	pos
ijassa-122	1	1	advances	advance	NOUN
ijassa-122	1	2	in	in	ADP
ijassa-122	1	3	systems	system	NOUN
ijassa-122	1	4	science	science	NOUN
ijassa-122	1	5	and	and	CCONJ
ijassa-122	1	6	applications	application	NOUN
ijassa-122	1	7	(	(	PUNCT
ijassa-122	1	8	2012	2012	NUM
ijassa-122	1	9	)	)	PUNCT
ijassa-122	1	10	vol.12	vol.12	NOUN
ijassa-122	1	11	no.4	no.4	PROPN
ijassa-122	1	12	399	399	NUM
ijassa-122	1	13	-	-	SYM
ijassa-122	1	14	405	405	NUM
ijassa-122	1	15	the	the	DET
ijassa-122	1	16	identification	identification	NOUN
ijassa-122	1	17	of	of	ADP
ijassa-122	1	18	outliers	outlier	NOUN
ijassa-122	1	19	in	in	ADP
ijassa-122	1	20	armax	armax	NOUN
ijassa-122	1	21	models	model	NOUN
ijassa-122	1	22	via	via	ADP
ijassa-122	1	23	genetic	genetic	ADJ
ijassa-122	1	24	algorithm	algorithm	NOUN
ijassa-122	1	25	ping	ping	NOUN
ijassa-122	1	26	chen1	chen1	PROPN
ijassa-122	1	27	and	and	CCONJ
ijassa-122	1	28	ying	ying	PROPN
ijassa-122	1	29	chen2	chen2	NOUN
ijassa-122	2	1	1department	1department	NUM
ijassa-122	2	2	of	of	ADP
ijassa-122	2	3	mathematics	mathematic	NOUN
ijassa-122	2	4	,	,	PUNCT
ijassa-122	2	5	southeast	southeast	ADJ
ijassa-122	2	6	university	university	NOUN
ijassa-122	2	7	,	,	PUNCT
ijassa-122	2	8	nanjing	nanjing	PROPN
ijassa-122	2	9	210096	210096	NUM
ijassa-122	2	10	,	,	PUNCT
ijassa-122	2	11	china	china	PROPN
ijassa-122	2	12	2department	2department	NUM
ijassa-122	2	13	of	of	ADP
ijassa-122	2	14	forensic	forensic	ADJ
ijassa-122	2	15	science	science	NOUN
ijassa-122	2	16	,	,	PUNCT
ijassa-122	2	17	jiangsu	jiangsu	PROPN
ijassa-122	2	18	police	police	PROPN
ijassa-122	2	19	institute	institute	PROPN
ijassa-122	2	20	,	,	PUNCT
ijassa-122	2	21	nanjing	nanjing	PROPN
ijassa-122	2	22	,	,	PUNCT
ijassa-122	2	23	210012	210012	NUM
ijassa-122	2	24	,	,	PUNCT
ijassa-122	2	25	china	china	PROPN
ijassa-122	2	26	abstract	abstract	ADJ
ijassa-122	2	27	this	this	DET
ijassa-122	2	28	paper	paper	NOUN
ijassa-122	2	29	proposes	propose	VERB
ijassa-122	2	30	a	a	DET
ijassa-122	2	31	procedure	procedure	NOUN
ijassa-122	2	32	to	to	PART
ijassa-122	2	33	identify	identify	VERB
ijassa-122	2	34	additive	additive	ADJ
ijassa-122	2	35	and	and	CCONJ
ijassa-122	2	36	innovational	innovational	ADJ
ijassa-122	2	37	outliers	outlier	NOUN
ijassa-122	2	38	by	by	ADP
ijassa-122	2	39	genetic	genetic	ADJ
ijassa-122	2	40	algorithm	algorithm	NOUN
ijassa-122	2	41	in	in	ADP
ijassa-122	2	42	autoregressive	autoregressive	ADJ
ijassa-122	2	43	moving	move	VERB
ijassa-122	2	44	average	average	NOUN
ijassa-122	2	45	with	with	ADP
ijassa-122	2	46	exogenous	exogenous	ADJ
ijassa-122	2	47	variable(armax	variable(armax	NOUN
ijassa-122	2	48	)	)	PUNCT
ijassa-122	2	49	time	time	NOUN
ijassa-122	2	50	series	series	NOUN
ijassa-122	2	51	models	model	NOUN
ijassa-122	2	52	.	.	PUNCT
ijassa-122	3	1	we	we	PRON
ijassa-122	3	2	use	use	VERB
ijassa-122	3	3	some	some	DET
ijassa-122	3	4	methods	method	NOUN
ijassa-122	3	5	to	to	PART
ijassa-122	3	6	delete	delete	VERB
ijassa-122	3	7	the	the	DET
ijassa-122	3	8	influence	influence	NOUN
ijassa-122	3	9	of	of	ADP
ijassa-122	3	10	input	input	NOUN
ijassa-122	3	11	process	process	NOUN
ijassa-122	3	12	in	in	ADP
ijassa-122	3	13	armax	armax	NOUN
ijassa-122	3	14	model	model	NOUN
ijassa-122	3	15	and	and	CCONJ
ijassa-122	3	16	then	then	ADV
ijassa-122	3	17	detect	detect	VERB
ijassa-122	3	18	outliers	outlier	NOUN
ijassa-122	3	19	in	in	ADP
ijassa-122	3	20	time	time	NOUN
ijassa-122	3	21	series	series	NOUN
ijassa-122	3	22	based	base	VERB
ijassa-122	3	23	on	on	ADP
ijassa-122	3	24	the	the	DET
ijassa-122	3	25	previous	previous	ADJ
ijassa-122	3	26	work	work	NOUN
ijassa-122	3	27	,	,	PUNCT
ijassa-122	3	28	which	which	PRON
ijassa-122	3	29	is	be	AUX
ijassa-122	3	30	an	an	DET
ijassa-122	3	31	improvement	improvement	NOUN
ijassa-122	3	32	and	and	CCONJ
ijassa-122	3	33	extension	extension	NOUN
ijassa-122	3	34	of	of	ADP
ijassa-122	3	35	the	the	DET
ijassa-122	3	36	detection	detection	NOUN
ijassa-122	3	37	method	method	NOUN
ijassa-122	3	38	on	on	ADP
ijassa-122	3	39	arma	arma	NOUN
ijassa-122	3	40	models	model	NOUN
ijassa-122	3	41	.	.	PUNCT
ijassa-122	4	1	empirical	empirical	ADJ
ijassa-122	4	2	and	and	CCONJ
ijassa-122	4	3	simulation	simulation	NOUN
ijassa-122	4	4	studies	study	NOUN
ijassa-122	4	5	show	show	VERB
ijassa-122	4	6	that	that	SCONJ
ijassa-122	4	7	the	the	DET
ijassa-122	4	8	proposed	propose	VERB
ijassa-122	4	9	procedure	procedure	NOUN
ijassa-122	4	10	is	be	AUX
ijassa-122	4	11	effective	effective	ADJ
ijassa-122	4	12	.	.	PUNCT
ijassa-122	5	1	keywords	keyword	VERB
ijassa-122	5	2	dynamic	dynamic	ADJ
ijassa-122	5	3	systems	system	NOUN
ijassa-122	5	4	,	,	PUNCT
ijassa-122	5	5	innovational	innovational	ADJ
ijassa-122	5	6	outliers	outlier	NOUN
ijassa-122	5	7	,	,	PUNCT
ijassa-122	5	8	armax	armax	NOUN
ijassa-122	5	9	model	model	NOUN
ijassa-122	5	10	,	,	PUNCT
ijassa-122	5	11	genetic	genetic	ADJ
ijassa-122	5	12	algorithm	algorithm	NOUN
ijassa-122	5	13	1	1	NUM
ijassa-122	5	14	introduction	introduction	NOUN
ijassa-122	5	15	outliers	outlier	NOUN
ijassa-122	5	16	in	in	ADP
ijassa-122	5	17	dynamic	dynamic	ADJ
ijassa-122	5	18	systems	system	NOUN
ijassa-122	5	19	or	or	CCONJ
ijassa-122	5	20	engineering	engineering	NOUN
ijassa-122	5	21	time	time	NOUN
ijassa-122	5	22	series	series	PROPN
ijassa-122	5	23	can	can	AUX
ijassa-122	5	24	have	have	VERB
ijassa-122	5	25	adverse	adverse	ADJ
ijassa-122	5	26	effects	effect	NOUN
ijassa-122	5	27	on	on	ADP
ijassa-122	5	28	model	model	NOUN
ijassa-122	5	29	identification	identification	NOUN
ijassa-122	5	30	and	and	CCONJ
ijassa-122	5	31	parameter	parameter	NOUN
ijassa-122	5	32	estimation	estimation	NOUN
ijassa-122	5	33	.	.	PUNCT
ijassa-122	6	1	several	several	ADJ
ijassa-122	6	2	procedures	procedure	NOUN
ijassa-122	6	3	are	be	AUX
ijassa-122	6	4	available	available	ADJ
ijassa-122	6	5	in	in	ADP
ijassa-122	6	6	literature	literature	NOUN
ijassa-122	6	7	to	to	PART
ijassa-122	6	8	handle	handle	VERB
ijassa-122	6	9	outliers	outlier	NOUN
ijassa-122	6	10	in	in	ADP
ijassa-122	6	11	a	a	DET
ijassa-122	6	12	time	time	NOUN
ijassa-122	6	13	series	series	NOUN
ijassa-122	6	14	.	.	PUNCT
ijassa-122	7	1	however	however	ADV
ijassa-122	7	2	,	,	PUNCT
ijassa-122	7	3	the	the	DET
ijassa-122	7	4	case	case	NOUN
ijassa-122	7	5	of	of	ADP
ijassa-122	7	6	multiple	multiple	ADJ
ijassa-122	7	7	additive	additive	ADJ
ijassa-122	7	8	outliers	outlier	NOUN
ijassa-122	7	9	and	and	CCONJ
ijassa-122	7	10	innovational	innovational	ADJ
ijassa-122	7	11	outliers	outlier	NOUN
ijassa-122	7	12	is	be	AUX
ijassa-122	7	13	very	very	ADV
ijassa-122	7	14	difficult	difficult	ADJ
ijassa-122	7	15	to	to	PART
ijassa-122	7	16	study	study	VERB
ijassa-122	7	17	because	because	SCONJ
ijassa-122	7	18	of	of	ADP
ijassa-122	7	19	the	the	DET
ijassa-122	7	20	great	great	ADJ
ijassa-122	7	21	number	number	NOUN
ijassa-122	7	22	of	of	ADP
ijassa-122	7	23	alternatives	alternative	NOUN
ijassa-122	7	24	and	and	CCONJ
ijassa-122	7	25	of	of	ADP
ijassa-122	7	26	the	the	DET
ijassa-122	7	27	masking	masking	NOUN
ijassa-122	7	28	and	and	CCONJ
ijassa-122	7	29	swamping	swamping	ADJ
ijassa-122	7	30	effects	effect	NOUN
ijassa-122	7	31	.	.	PUNCT
ijassa-122	8	1	compared	compare	VERB
ijassa-122	8	2	to	to	ADP
ijassa-122	8	3	other	other	ADJ
ijassa-122	8	4	search	search	NOUN
ijassa-122	8	5	algorithms	algorithm	NOUN
ijassa-122	8	6	,	,	PUNCT
ijassa-122	8	7	genetic	genetic	ADJ
ijassa-122	8	8	algorithms	algorithm	NOUN
ijassa-122	8	9	allow	allow	VERB
ijassa-122	8	10	many	many	ADJ
ijassa-122	8	11	candidate	candidate	NOUN
ijassa-122	8	12	solutions	solution	NOUN
ijassa-122	8	13	to	to	PART
ijassa-122	8	14	be	be	AUX
ijassa-122	8	15	considered	consider	VERB
ijassa-122	8	16	simultaneously	simultaneously	ADV
ijassa-122	8	17	at	at	ADP
ijassa-122	8	18	each	each	DET
ijassa-122	8	19	step	step	NOUN
ijassa-122	8	20	.	.	PUNCT
ijassa-122	9	1	baragona	baragona	PROPN
ijassa-122	9	2	et	et	PROPN
ijassa-122	9	3	al.[1	al.[1	PROPN
ijassa-122	9	4	]	]	PUNCT
ijassa-122	9	5	showed	show	VERB
ijassa-122	9	6	how	how	SCONJ
ijassa-122	9	7	to	to	PART
ijassa-122	9	8	use	use	VERB
ijassa-122	9	9	genetic	genetic	ADJ
ijassa-122	9	10	algorithms	algorithm	NOUN
ijassa-122	9	11	for	for	ADP
ijassa-122	9	12	outlier	outlier	NOUN
ijassa-122	9	13	detection	detection	NOUN
ijassa-122	9	14	and	and	CCONJ
ijassa-122	9	15	classification	classification	NOUN
ijassa-122	9	16	in	in	ADP
ijassa-122	9	17	arma	arma	PROPN
ijassa-122	9	18	series	series	PROPN
ijassa-122	9	19	.	.	PUNCT
ijassa-122	10	1	peña	peña	PROPN
ijassa-122	10	2	and	and	CCONJ
ijassa-122	10	3	sánchez[2	sánchez[2	PROPN
ijassa-122	10	4	]	]	PUNCT
ijassa-122	10	5	presented	present	VERB
ijassa-122	10	6	a	a	DET
ijassa-122	10	7	new	new	ADJ
ijassa-122	10	8	procedure	procedure	NOUN
ijassa-122	10	9	for	for	ADP
ijassa-122	10	10	multifold	multifold	ADJ
ijassa-122	10	11	predictive	predictive	ADJ
ijassa-122	10	12	validation	validation	NOUN
ijassa-122	10	13	in	in	ADP
ijassa-122	10	14	armax	armax	NOUN
ijassa-122	10	15	models	model	NOUN
ijassa-122	10	16	.	.	PUNCT
ijassa-122	11	1	also	also	ADV
ijassa-122	11	2	,	,	PUNCT
ijassa-122	11	3	chen	chen	PROPN
ijassa-122	11	4	et	et	PROPN
ijassa-122	11	5	al.[3	al.[3	PROPN
ijassa-122	11	6	]	]	PUNCT
ijassa-122	11	7	and	and	CCONJ
ijassa-122	11	8	chen	chen	PROPN
ijassa-122	11	9	et	et	PROPN
ijassa-122	11	10	al.[4	al.[4	PROPN
ijassa-122	11	11	]	]	PUNCT
ijassa-122	11	12	developed	develop	VERB
ijassa-122	11	13	some	some	DET
ijassa-122	11	14	methods	method	NOUN
ijassa-122	11	15	for	for	ADP
ijassa-122	11	16	detecting	detect	VERB
ijassa-122	11	17	outliers	outlier	NOUN
ijassa-122	11	18	,	,	PUNCT
ijassa-122	11	19	change	change	NOUN
ijassa-122	11	20	point	point	NOUN
ijassa-122	11	21	and	and	CCONJ
ijassa-122	11	22	outlier	outlier	NOUN
ijassa-122	11	23	patches	patch	NOUN
ijassa-122	11	24	in	in	ADP
ijassa-122	11	25	bilinear	bilinear	PROPN
ijassa-122	11	26	time	time	NOUN
ijassa-122	11	27	series	series	PROPN
ijassa-122	11	28	models	model	NOUN
ijassa-122	11	29	.	.	PUNCT
ijassa-122	12	1	on	on	ADP
ijassa-122	12	2	the	the	DET
ijassa-122	12	3	other	other	ADJ
ijassa-122	12	4	hand	hand	NOUN
ijassa-122	12	5	,	,	PUNCT
ijassa-122	12	6	huang	huang	PROPN
ijassa-122	12	7	et	et	PROPN
ijassa-122	12	8	al.[5	al.[5	PROPN
ijassa-122	12	9	]	]	PUNCT
ijassa-122	12	10	discussed	discuss	VERB
ijassa-122	12	11	the	the	DET
ijassa-122	12	12	improved	improve	VERB
ijassa-122	12	13	genetic	genetic	ADJ
ijassa-122	12	14	algorithm	algorithm	NOUN
ijassa-122	12	15	for	for	ADP
ijassa-122	12	16	vehicle	vehicle	NOUN
ijassa-122	12	17	routing	routing	NOUN
ijassa-122	12	18	problem	problem	NOUN
ijassa-122	12	19	with	with	ADP
ijassa-122	12	20	time	time	NOUN
ijassa-122	12	21	windows	window	NOUN
ijassa-122	12	22	.	.	PUNCT
ijassa-122	13	1	in	in	ADP
ijassa-122	13	2	this	this	DET
ijassa-122	13	3	paper	paper	NOUN
ijassa-122	13	4	,	,	PUNCT
ijassa-122	13	5	a	a	DET
ijassa-122	13	6	genetic	genetic	ADJ
ijassa-122	13	7	algorithm	algorithm	NOUN
ijassa-122	13	8	is	be	AUX
ijassa-122	13	9	proposed	propose	VERB
ijassa-122	13	10	to	to	PART
ijassa-122	13	11	identify	identify	VERB
ijassa-122	13	12	additive	additive	ADJ
ijassa-122	13	13	and	and	CCONJ
ijassa-122	13	14	innovational	innovational	ADJ
ijassa-122	13	15	outliers	outlier	NOUN
ijassa-122	13	16	in	in	ADP
ijassa-122	13	17	armax	armax	ADJ
ijassa-122	13	18	series	series	NOUN
ijassa-122	13	19	.	.	PUNCT
ijassa-122	14	1	we	we	PRON
ijassa-122	14	2	are	be	AUX
ijassa-122	14	3	using	use	VERB
ijassa-122	14	4	the	the	DET
ijassa-122	14	5	standard	standard	ADJ
ijassa-122	14	6	genetic	genetic	ADJ
ijassa-122	14	7	algorithm	algorithm	NOUN
ijassa-122	14	8	with	with	ADP
ijassa-122	14	9	complete	complete	ADJ
ijassa-122	14	10	replacement	replacement	NOUN
ijassa-122	14	11	of	of	ADP
ijassa-122	14	12	the	the	DET
ijassa-122	14	13	past	past	ADJ
ijassa-122	14	14	population	population	NOUN
ijassa-122	14	15	and	and	CCONJ
ijassa-122	14	16	elitist	elitist	ADJ
ijassa-122	14	17	strategy	strategy	NOUN
ijassa-122	14	18	.	.	PUNCT
ijassa-122	15	1	the	the	DET
ijassa-122	15	2	relationship	relationship	NOUN
ijassa-122	15	3	between	between	ADP
ijassa-122	15	4	inverse	inverse	NOUN
ijassa-122	15	5	correlations	correlation	NOUN
ijassa-122	15	6	and	and	CCONJ
ijassa-122	15	7	outliers	outlier	NOUN
ijassa-122	15	8	is	be	AUX
ijassa-122	15	9	helpful	helpful	ADJ
ijassa-122	15	10	to	to	PART
ijassa-122	15	11	simplify	simplify	VERB
ijassa-122	15	12	the	the	DET
ijassa-122	15	13	fitness	fitness	NOUN
ijassa-122	15	14	function	function	NOUN
ijassa-122	15	15	,	,	PUNCT
ijassa-122	15	16	which	which	PRON
ijassa-122	15	17	may	may	AUX
ijassa-122	15	18	be	be	AUX
ijassa-122	15	19	quickly	quickly	ADV
ijassa-122	15	20	computed	compute	VERB
ijassa-122	15	21	by	by	ADP
ijassa-122	15	22	trench	trench	NOUN
ijassa-122	15	23	’s	’s	PART
ijassa-122	15	24	algorithm	algorithm	NOUN
ijassa-122	15	25	.	.	PUNCT
ijassa-122	16	1	for	for	ADP
ijassa-122	16	2	the	the	DET
ijassa-122	16	3	case	case	NOUN
ijassa-122	16	4	of	of	ADP
ijassa-122	16	5	large	large	ADJ
ijassa-122	16	6	series	series	NOUN
ijassa-122	16	7	,	,	PUNCT
ijassa-122	16	8	it	it	PRON
ijassa-122	16	9	is	be	AUX
ijassa-122	16	10	better	well	ADJ
ijassa-122	16	11	to	to	PART
ijassa-122	16	12	divide	divide	VERB
ijassa-122	16	13	the	the	DET
ijassa-122	16	14	series	series	NOUN
ijassa-122	16	15	into	into	ADP
ijassa-122	16	16	several	several	ADJ
ijassa-122	16	17	parts	part	NOUN
ijassa-122	16	18	so	so	SCONJ
ijassa-122	16	19	that	that	SCONJ
ijassa-122	16	20	the	the	DET
ijassa-122	16	21	two	two	NUM
ijassa-122	16	22	neighbor	neighbor	NOUN
ijassa-122	16	23	subseries	subserie	NOUN
ijassa-122	16	24	share	share	VERB
ijassa-122	16	25	a	a	DET
ijassa-122	16	26	length	length	NOUN
ijassa-122	16	27	of	of	ADP
ijassa-122	16	28	same	same	ADJ
ijassa-122	16	29	data	datum	NOUN
ijassa-122	16	30	,	,	PUNCT
ijassa-122	16	31	and	and	CCONJ
ijassa-122	16	32	then	then	ADV
ijassa-122	16	33	to	to	PART
ijassa-122	16	34	detect	detect	VERB
ijassa-122	16	35	the	the	DET
ijassa-122	16	36	subseries	subserie	NOUN
ijassa-122	16	37	respectively	respectively	ADV
ijassa-122	16	38	.	.	PUNCT
ijassa-122	17	1	thus	thus	ADV
ijassa-122	17	2	,	,	PUNCT
ijassa-122	17	3	the	the	DET
ijassa-122	17	4	problem	problem	NOUN
ijassa-122	17	5	of	of	ADP
ijassa-122	17	6	large	large	ADJ
ijassa-122	17	7	population	population	NOUN
ijassa-122	17	8	in	in	ADP
ijassa-122	17	9	genetic	genetic	ADJ
ijassa-122	17	10	algorithms	algorithm	NOUN
ijassa-122	17	11	has	have	AUX
ijassa-122	17	12	been	be	AUX
ijassa-122	17	13	avoided	avoid	VERB
ijassa-122	17	14	,	,	PUNCT
ijassa-122	17	15	as	as	ADV
ijassa-122	17	16	well	well	ADV
ijassa-122	17	17	as	as	ADP
ijassa-122	17	18	400	400	NUM
ijassa-122	17	19	ping	ping	NOUN
ijassa-122	17	20	chen	chen	PROPN
ijassa-122	17	21	:	:	PUNCT
ijassa-122	17	22	the	the	DET
ijassa-122	17	23	identification	identification	NOUN
ijassa-122	17	24	of	of	ADP
ijassa-122	17	25	outliers	outlier	NOUN
ijassa-122	17	26	in	in	ADP
ijassa-122	17	27	armax	armax	NOUN
ijassa-122	17	28	models	model	NOUN
ijassa-122	17	29	via	via	ADP
ijassa-122	17	30	genetic	genetic	ADJ
ijassa-122	17	31	algorithm	algorithm	NOUN
ijassa-122	17	32	the	the	DET
ijassa-122	17	33	loss	loss	NOUN
ijassa-122	17	34	of	of	ADP
ijassa-122	17	35	outliers	outlier	NOUN
ijassa-122	17	36	around	around	ADP
ijassa-122	17	37	the	the	DET
ijassa-122	17	38	cut	cut	NOUN
ijassa-122	17	39	point	point	NOUN
ijassa-122	17	40	.	.	PUNCT
ijassa-122	18	1	at	at	ADP
ijassa-122	18	2	last	last	ADJ
ijassa-122	18	3	,	,	PUNCT
ijassa-122	18	4	simulation	simulation	NOUN
ijassa-122	18	5	studies	study	NOUN
ijassa-122	18	6	are	be	AUX
ijassa-122	18	7	carried	carry	VERB
ijassa-122	18	8	out	out	ADP
ijassa-122	18	9	,	,	PUNCT
ijassa-122	18	10	which	which	PRON
ijassa-122	18	11	show	show	VERB
ijassa-122	18	12	promising	promising	ADJ
ijassa-122	18	13	results	result	NOUN
ijassa-122	18	14	.	.	PUNCT
ijassa-122	19	1	2	2	NUM
ijassa-122	19	2	genetic	genetic	ADJ
ijassa-122	19	3	algorithm	algorithm	NOUN
ijassa-122	19	4	and	and	CCONJ
ijassa-122	19	5	outliers	outlier	NOUN
ijassa-122	19	6	models	model	NOUN
ijassa-122	19	7	in	in	ADP
ijassa-122	19	8	armax	armax	ADJ
ijassa-122	19	9	series	series	NOUN
ijassa-122	19	10	the	the	DET
ijassa-122	19	11	genetic	genetic	ADJ
ijassa-122	19	12	algorithm(ga	algorithm(ga	NOUN
ijassa-122	19	13	)	)	PUNCT
ijassa-122	19	14	is	be	AUX
ijassa-122	19	15	known	know	VERB
ijassa-122	19	16	to	to	PART
ijassa-122	19	17	be	be	AUX
ijassa-122	19	18	able	able	ADJ
ijassa-122	19	19	to	to	PART
ijassa-122	19	20	provide	provide	VERB
ijassa-122	19	21	us	we	PRON
ijassa-122	19	22	with	with	ADP
ijassa-122	19	23	a	a	DET
ijassa-122	19	24	powerful	powerful	ADJ
ijassa-122	19	25	optimization	optimization	NOUN
ijassa-122	19	26	tool	tool	NOUN
ijassa-122	19	27	when	when	SCONJ
ijassa-122	19	28	the	the	DET
ijassa-122	19	29	solution	solution	NOUN
ijassa-122	19	30	space	space	NOUN
ijassa-122	19	31	happens	happen	VERB
ijassa-122	19	32	to	to	PART
ijassa-122	19	33	be	be	AUX
ijassa-122	19	34	both	both	CCONJ
ijassa-122	19	35	discrete	discrete	ADJ
ijassa-122	19	36	and	and	CCONJ
ijassa-122	19	37	large	large	ADJ
ijassa-122	19	38	,	,	PUNCT
ijassa-122	19	39	and	and	CCONJ
ijassa-122	19	40	the	the	DET
ijassa-122	19	41	objective	objective	ADJ
ijassa-122	19	42	function	function	NOUN
ijassa-122	19	43	does	do	AUX
ijassa-122	19	44	not	not	PART
ijassa-122	19	45	fulfill	fulfill	VERB
ijassa-122	19	46	the	the	DET
ijassa-122	19	47	usual	usual	ADJ
ijassa-122	19	48	regularity	regularity	NOUN
ijassa-122	19	49	requirements.the	requirements.the	DET
ijassa-122	19	50	key	key	ADJ
ijassa-122	19	51	feature	feature	NOUN
ijassa-122	19	52	of	of	ADP
ijassa-122	19	53	a	a	DET
ijassa-122	19	54	ga	ga	PROPN
ijassa-122	19	55	is	be	AUX
ijassa-122	19	56	the	the	DET
ijassa-122	19	57	manipulation	manipulation	NOUN
ijassa-122	19	58	of	of	ADP
ijassa-122	19	59	a	a	DET
ijassa-122	19	60	population	population	NOUN
ijassa-122	19	61	whose	whose	DET
ijassa-122	19	62	individuals	individual	NOUN
ijassa-122	19	63	are	be	AUX
ijassa-122	19	64	characterized	characterize	VERB
ijassa-122	19	65	by	by	ADP
ijassa-122	19	66	possessing	possess	VERB
ijassa-122	19	67	a	a	DET
ijassa-122	19	68	chromosome	chromosome	NOUN
ijassa-122	19	69	.	.	PUNCT
ijassa-122	20	1	this	this	DET
ijassa-122	20	2	latter	latter	ADJ
ijassa-122	20	3	can	can	AUX
ijassa-122	20	4	be	be	AUX
ijassa-122	20	5	coded	code	VERB
ijassa-122	20	6	as	as	ADP
ijassa-122	20	7	a	a	DET
ijassa-122	20	8	string	string	NOUN
ijassa-122	20	9	of	of	ADP
ijassa-122	20	10	characters	character	NOUN
ijassa-122	20	11	of	of	ADP
ijassa-122	20	12	given	give	VERB
ijassa-122	20	13	length	length	NOUN
ijassa-122	20	14	.	.	PUNCT
ijassa-122	21	1	each	each	DET
ijassa-122	21	2	string	string	NOUN
ijassa-122	21	3	represents	represent	VERB
ijassa-122	21	4	a	a	DET
ijassa-122	21	5	feasible	feasible	ADJ
ijassa-122	21	6	solution	solution	NOUN
ijassa-122	21	7	to	to	ADP
ijassa-122	21	8	the	the	DET
ijassa-122	21	9	optimization	optimization	NOUN
ijassa-122	21	10	problem	problem	NOUN
ijassa-122	21	11	.	.	PUNCT
ijassa-122	22	1	the	the	DET
ijassa-122	22	2	link	link	NOUN
ijassa-122	22	3	between	between	ADP
ijassa-122	22	4	the	the	DET
ijassa-122	22	5	ga	ga	PROPN
ijassa-122	22	6	and	and	CCONJ
ijassa-122	22	7	the	the	DET
ijassa-122	22	8	problem	problem	NOUN
ijassa-122	22	9	at	at	ADP
ijassa-122	22	10	hand	hand	NOUN
ijassa-122	22	11	is	be	AUX
ijassa-122	22	12	provided	provide	VERB
ijassa-122	22	13	by	by	ADP
ijassa-122	22	14	the	the	DET
ijassa-122	22	15	fitness	fitness	NOUN
ijassa-122	22	16	function	function	NOUN
ijassa-122	22	17	(	(	PUNCT
ijassa-122	22	18	ff	ff	NOUN
ijassa-122	22	19	)	)	PUNCT
ijassa-122	22	20	.	.	PUNCT
ijassa-122	23	1	the	the	DET
ijassa-122	23	2	ff	ff	NOUN
ijassa-122	23	3	establishes	establish	VERB
ijassa-122	23	4	a	a	DET
ijassa-122	23	5	mapping	mapping	NOUN
ijassa-122	23	6	from	from	ADP
ijassa-122	23	7	the	the	DET
ijassa-122	23	8	chromosomes	chromosome	NOUN
ijassa-122	23	9	to	to	ADP
ijassa-122	23	10	some	some	DET
ijassa-122	23	11	set	set	NOUN
ijassa-122	23	12	of	of	ADP
ijassa-122	23	13	real	real	ADJ
ijassa-122	23	14	numbers	number	NOUN
ijassa-122	23	15	.	.	PUNCT
ijassa-122	24	1	the	the	PRON
ijassa-122	24	2	greater	great	ADJ
ijassa-122	24	3	the	the	DET
ijassa-122	24	4	ff	ff	NOUN
ijassa-122	24	5	is	be	AUX
ijassa-122	24	6	,	,	PUNCT
ijassa-122	24	7	the	the	PRON
ijassa-122	24	8	better	well	ADJ
ijassa-122	24	9	the	the	DET
ijassa-122	24	10	adaptation	adaptation	NOUN
ijassa-122	24	11	of	of	ADP
ijassa-122	24	12	the	the	DET
ijassa-122	24	13	individual	individual	NOUN
ijassa-122	24	14	.	.	PUNCT
ijassa-122	25	1	the	the	DET
ijassa-122	25	2	procedure	procedure	NOUN
ijassa-122	25	3	is	be	AUX
ijassa-122	25	4	iterative	iterative	NOUN
ijassa-122	25	5	.	.	PUNCT
ijassa-122	26	1	it	it	PRON
ijassa-122	26	2	makes	make	VERB
ijassa-122	26	3	use	use	NOUN
ijassa-122	26	4	of	of	ADP
ijassa-122	26	5	three	three	NUM
ijassa-122	26	6	evolutionary	evolutionary	ADJ
ijassa-122	26	7	operators	operator	NOUN
ijassa-122	26	8	:	:	PUNCT
ijassa-122	26	9	reproduction	reproduction	NOUN
ijassa-122	26	10	,	,	PUNCT
ijassa-122	26	11	crossover	crossover	NOUN
ijassa-122	26	12	and	and	CCONJ
ijassa-122	26	13	mutation	mutation	NOUN
ijassa-122	26	14	.	.	PUNCT
ijassa-122	27	1	an	an	DET
ijassa-122	27	2	arma	arma	PROPN
ijassa-122	27	3	model	model	NOUN
ijassa-122	27	4	with	with	ADP
ijassa-122	27	5	input	input	NOUN
ijassa-122	27	6	process	process	NOUN
ijassa-122	27	7	is	be	AUX
ijassa-122	27	8	called	call	VERB
ijassa-122	27	9	armax	armax	ADJ
ijassa-122	27	10	model	model	NOUN
ijassa-122	27	11	,	,	PUNCT
ijassa-122	27	12	which	which	PRON
ijassa-122	27	13	is	be	AUX
ijassa-122	27	14	defined	define	VERB
ijassa-122	27	15	as	as	ADP
ijassa-122	27	16	zt	zt	PROPN
ijassa-122	27	17	=	=	PROPN
ijassa-122	27	18	d∑	d∑	PROPN
ijassa-122	27	19	i=1	i=1	PROPN
ijassa-122	27	20	υi(b)xi	υi(b)xi	NOUN
ijassa-122	27	21	,	,	PUNCT
ijassa-122	27	22	t	t	PROPN
ijassa-122	28	1	+	+	CCONJ
ijassa-122	28	2	nt	not	PART
ijassa-122	28	3	,	,	PUNCT
ijassa-122	28	4	where	where	SCONJ
ijassa-122	28	5	υi(b	υi(b	PUNCT
ijassa-122	28	6	)	)	PUNCT
ijassa-122	29	1	=	=	SYM
ijassa-122	30	1	(	(	PUNCT
ijassa-122	30	2	δ−1	δ−1	PROPN
ijassa-122	30	3	i	i	PRON
ijassa-122	30	4	(	(	PUNCT
ijassa-122	30	5	b	b	NOUN
ijassa-122	30	6	)	)	PUNCT
ijassa-122	30	7	·	·	PUNCT
ijassa-122	31	1	ωi(b))bki	ωi(b))bki	NUM
ijassa-122	31	2	is	be	AUX
ijassa-122	31	3	the	the	DET
ijassa-122	31	4	transfer	transfer	NOUN
ijassa-122	31	5	function	function	NOUN
ijassa-122	31	6	of	of	ADP
ijassa-122	31	7	ith	ith	PROPN
ijassa-122	31	8	input	input	NOUN
ijassa-122	31	9	process	process	NOUN
ijassa-122	31	10	,	,	PUNCT
ijassa-122	31	11	nt	not	PART
ijassa-122	31	12	=	=	PRON
ijassa-122	31	13	(	(	PUNCT
ijassa-122	31	14	θ(b)/ϕ(b))εt	θ(b)/ϕ(b))εt	PROPN
ijassa-122	31	15	is	be	AUX
ijassa-122	31	16	noise	noise	NOUN
ijassa-122	31	17	process	process	NOUN
ijassa-122	31	18	.	.	PUNCT
ijassa-122	32	1	{	{	PUNCT
ijassa-122	32	2	zt	zt	PROPN
ijassa-122	32	3	}	}	PUNCT
ijassa-122	32	4	is	be	AUX
ijassa-122	32	5	called	call	VERB
ijassa-122	32	6	response	response	NOUN
ijassa-122	32	7	process	process	NOUN
ijassa-122	32	8	.	.	PUNCT
ijassa-122	33	1	and	and	CCONJ
ijassa-122	33	2	xi	xi	PROPN
ijassa-122	33	3	,	,	PUNCT
ijassa-122	33	4	t	t	PROPN
ijassa-122	33	5	denotes	denote	VERB
ijassa-122	33	6	the	the	DET
ijassa-122	33	7	ith	ith	PROPN
ijassa-122	33	8	input	input	NOUN
ijassa-122	33	9	process	process	NOUN
ijassa-122	33	10	or	or	CCONJ
ijassa-122	33	11	the	the	DET
ijassa-122	33	12	difference	difference	NOUN
ijassa-122	33	13	of	of	ADP
ijassa-122	33	14	ith	ith	PROPN
ijassa-122	33	15	input	input	NOUN
ijassa-122	33	16	process	process	NOUN
ijassa-122	33	17	at	at	ADP
ijassa-122	33	18	time	time	NOUN
ijassa-122	33	19	t	t	PROPN
ijassa-122	33	20	,	,	PUNCT
ijassa-122	33	21	ki	ki	PROPN
ijassa-122	33	22	presents	present	VERB
ijassa-122	33	23	the	the	DET
ijassa-122	33	24	influence	influence	NOUN
ijassa-122	33	25	’s	’s	PART
ijassa-122	33	26	time	time	NOUN
ijassa-122	33	27	delay	delay	NOUN
ijassa-122	33	28	of	of	ADP
ijassa-122	33	29	ith	ith	PROPN
ijassa-122	33	30	input	input	NOUN
ijassa-122	33	31	process	process	NOUN
ijassa-122	33	32	,	,	PUNCT
ijassa-122	33	33	εt	εt	PROPN
ijassa-122	33	34	is	be	AUX
ijassa-122	33	35	normal	normal	ADJ
ijassa-122	33	36	white	white	ADJ
ijassa-122	33	37	nose	nose	NOUN
ijassa-122	33	38	process	process	NOUN
ijassa-122	33	39	.	.	PUNCT
ijassa-122	34	1	θ(b	θ(b	NOUN
ijassa-122	34	2	)	)	PUNCT
ijassa-122	34	3	=	=	SYM
ijassa-122	34	4	1	1	NUM
ijassa-122	35	1	+	+	NUM
ijassa-122	35	2	θ1b	θ1b	NOUN
ijassa-122	35	3	+	+	X
ijassa-122	35	4	·	·	PUNCT
ijassa-122	35	5	·	·	PUNCT
ijassa-122	35	6	·	·	PUNCT
ijassa-122	35	7	+	+	NUM
ijassa-122	35	8	θqb	θqb	NOUN
ijassa-122	35	9	q	q	X
ijassa-122	35	10	,	,	PUNCT
ijassa-122	35	11	ϕ(b	ϕ(b	PROPN
ijassa-122	35	12	)	)	PUNCT
ijassa-122	35	13	=	=	SYM
ijassa-122	36	1	1−	1−	NUM
ijassa-122	36	2	ϕ1b	ϕ1b	SYM
ijassa-122	36	3	−	−	PROPN
ijassa-122	36	4	·	·	PUNCT
ijassa-122	36	5	·	·	PUNCT
ijassa-122	36	6	·	·	PUNCT
ijassa-122	37	1	−	−	NOUN
ijassa-122	38	1	ϕpb	ϕpb	X
ijassa-122	38	2	p	p	NOUN
ijassa-122	38	3	where	where	SCONJ
ijassa-122	38	4	b	b	NOUN
ijassa-122	38	5	is	be	AUX
ijassa-122	38	6	the	the	DET
ijassa-122	38	7	backshift	backshift	NOUN
ijassa-122	38	8	operator	operator	NOUN
ijassa-122	38	9	.	.	PUNCT
ijassa-122	39	1	when	when	SCONJ
ijassa-122	39	2	υi(b	υi(b	ADV
ijassa-122	39	3	)	)	PUNCT
ijassa-122	39	4	=	=	SYM
ijassa-122	40	1	0	0	NUM
ijassa-122	40	2	,	,	PUNCT
ijassa-122	40	3	i	i	PRON
ijassa-122	40	4	=	=	NOUN
ijassa-122	40	5	1	1	NUM
ijassa-122	40	6	,	,	PUNCT
ijassa-122	40	7	·	·	PUNCT
ijassa-122	40	8	·	·	PUNCT
ijassa-122	40	9	·	·	PUNCT
ijassa-122	40	10	,	,	PUNCT
ijassa-122	40	11	d	d	X
ijassa-122	40	12	,	,	PUNCT
ijassa-122	40	13	it	it	PRON
ijassa-122	40	14	is	be	AUX
ijassa-122	40	15	arma	arma	PROPN
ijassa-122	40	16	model	model	NOUN
ijassa-122	40	17	,	,	PUNCT
ijassa-122	40	18	when	when	SCONJ
ijassa-122	40	19	some	some	DET
ijassa-122	40	20	υi(b	υi(b	NUM
ijassa-122	40	21	)	)	PUNCT
ijassa-122	40	22	is	be	AUX
ijassa-122	40	23	nonzero	nonzero	X
ijassa-122	40	24	constant	constant	ADJ
ijassa-122	40	25	,	,	PUNCT
ijassa-122	40	26	i	i	NOUN
ijassa-122	40	27	=	=	NOUN
ijassa-122	40	28	1	1	NUM
ijassa-122	40	29	,	,	PUNCT
ijassa-122	40	30	·	·	PUNCT
ijassa-122	40	31	·	·	PUNCT
ijassa-122	40	32	·	·	PUNCT
ijassa-122	40	33	,	,	PUNCT
ijassa-122	40	34	d	d	X
ijassa-122	40	35	,	,	PUNCT
ijassa-122	40	36	it	it	PRON
ijassa-122	40	37	is	be	AUX
ijassa-122	40	38	regression	regression	NOUN
ijassa-122	40	39	model	model	NOUN
ijassa-122	40	40	with	with	ADP
ijassa-122	40	41	arma	arma	PROPN
ijassa-122	40	42	error	error	NOUN
ijassa-122	40	43	.	.	PUNCT
ijassa-122	41	1	(	(	PUNCT
ijassa-122	41	2	1	1	X
ijassa-122	41	3	)	)	PUNCT
ijassa-122	41	4	additive	additive	ADJ
ijassa-122	41	5	outliers(ao	outliers(ao	NOUN
ijassa-122	41	6	)	)	PUNCT
ijassa-122	41	7	model	model	NOUN
ijassa-122	41	8	:	:	PUNCT
ijassa-122	41	9	suppose	suppose	VERB
ijassa-122	41	10	that	that	SCONJ
ijassa-122	41	11	only	only	ADV
ijassa-122	41	12	the	the	DET
ijassa-122	41	13	jth	jth	PROPN
ijassa-122	41	14	point	point	PROPN
ijassa-122	41	15	zj	zj	PROPN
ijassa-122	41	16	be	be	AUX
ijassa-122	41	17	ao	ao	PROPN
ijassa-122	41	18	,	,	PUNCT
ijassa-122	41	19	whose	whose	DET
ijassa-122	41	20	influence	influence	NOUN
ijassa-122	41	21	magnitude	magnitude	NOUN
ijassa-122	41	22	is	be	AUX
ijassa-122	41	23	wtj	wtj	PROPN
ijassa-122	41	24	,	,	PUNCT
ijassa-122	41	25	then	then	ADV
ijassa-122	41	26	we	we	PRON
ijassa-122	41	27	have	have	VERB
ijassa-122	41	28	zt	zt	PROPN
ijassa-122	41	29	=	=	SYM
ijassa-122	41	30	d∑	d∑	PROPN
ijassa-122	41	31	i=1	i=1	PROPN
ijassa-122	41	32	υi(b)xi	υi(b)xi	NOUN
ijassa-122	41	33	,	,	PUNCT
ijassa-122	41	34	t	t	PROPN
ijassa-122	41	35	+	+	CCONJ
ijassa-122	41	36	wtjδt	wtjδt	ADV
ijassa-122	41	37	,	,	PUNCT
ijassa-122	41	38	tj	tj	PROPN
ijassa-122	41	39	+	+	CCONJ
ijassa-122	41	40	θ(b	θ(b	NOUN
ijassa-122	41	41	)	)	PUNCT
ijassa-122	41	42	ϕ(b	ϕ(b	PROPN
ijassa-122	41	43	)	)	PUNCT
ijassa-122	42	1	εt	εt	PROPN
ijassa-122	42	2	where	where	SCONJ
ijassa-122	42	3	δt	δt	X
ijassa-122	42	4	,	,	PUNCT
ijassa-122	42	5	tj	tj	PROPN
ijassa-122	42	6	is	be	AUX
ijassa-122	42	7	kronecker	kronecker	NOUN
ijassa-122	42	8	symbol	symbol	NOUN
ijassa-122	42	9	:	:	PUNCT
ijassa-122	42	10	if	if	SCONJ
ijassa-122	42	11	t	t	PROPN
ijassa-122	42	12	=	=	SYM
ijassa-122	42	13	tj	tj	PROPN
ijassa-122	42	14	,	,	PUNCT
ijassa-122	42	15	then	then	ADV
ijassa-122	42	16	δt	δt	NOUN
ijassa-122	42	17	,	,	PUNCT
ijassa-122	42	18	tj	tj	NOUN
ijassa-122	42	19	=	=	SYM
ijassa-122	42	20	1	1	NUM
ijassa-122	42	21	,	,	PUNCT
ijassa-122	42	22	else	else	ADV
ijassa-122	42	23	δt	δt	NOUN
ijassa-122	42	24	,	,	PUNCT
ijassa-122	42	25	tj	tj	NOUN
ijassa-122	42	26	=	=	SYM
ijassa-122	42	27	0	0	PROPN
ijassa-122	42	28	.	.	PUNCT
ijassa-122	43	1	(	(	PUNCT
ijassa-122	43	2	2	2	X
ijassa-122	43	3	)	)	PUNCT
ijassa-122	43	4	innovational	innovational	ADJ
ijassa-122	43	5	outliers(io	outliers(io	PROPN
ijassa-122	43	6	)	)	PUNCT
ijassa-122	43	7	model	model	NOUN
ijassa-122	43	8	suppose	suppose	VERB
ijassa-122	43	9	that	that	SCONJ
ijassa-122	43	10	only	only	ADV
ijassa-122	43	11	the	the	DET
ijassa-122	43	12	jth	jth	PROPN
ijassa-122	43	13	point	point	PROPN
ijassa-122	43	14	zj	zj	AUX
ijassa-122	43	15	be	be	AUX
ijassa-122	43	16	io	io	NOUN
ijassa-122	43	17	,	,	PUNCT
ijassa-122	43	18	whose	whose	DET
ijassa-122	43	19	influence	influence	NOUN
ijassa-122	43	20	magnitude	magnitude	NOUN
ijassa-122	43	21	is	be	AUX
ijassa-122	43	22	wtj	wtj	PROPN
ijassa-122	43	23	,	,	PUNCT
ijassa-122	43	24	advances	advance	NOUN
ijassa-122	43	25	in	in	ADP
ijassa-122	43	26	systems	system	NOUN
ijassa-122	43	27	science	science	NOUN
ijassa-122	43	28	and	and	CCONJ
ijassa-122	43	29	applications	application	NOUN
ijassa-122	43	30	(	(	PUNCT
ijassa-122	43	31	2012	2012	NUM
ijassa-122	43	32	)	)	PUNCT
ijassa-122	43	33	vol.12	vol.12	NOUN
ijassa-122	43	34	no.4	no.4	PROPN
ijassa-122	43	35	401	401	NUM
ijassa-122	43	36	then	then	ADV
ijassa-122	43	37	we	we	PRON
ijassa-122	43	38	have	have	VERB
ijassa-122	43	39	zt	zt	PROPN
ijassa-122	43	40	=	=	SYM
ijassa-122	43	41	d∑	d∑	PROPN
ijassa-122	43	42	i=1	i=1	PROPN
ijassa-122	43	43	υi(b)xi	υi(b)xi	NOUN
ijassa-122	43	44	,	,	PUNCT
ijassa-122	43	45	t	t	PROPN
ijassa-122	43	46	+	+	CCONJ
ijassa-122	43	47	θ(b	θ(b	NOUN
ijassa-122	43	48	)	)	PUNCT
ijassa-122	43	49	ϕ(b	ϕ(b	PROPN
ijassa-122	43	50	)	)	PUNCT
ijassa-122	44	1	(	(	PUNCT
ijassa-122	44	2	εt	εt	PROPN
ijassa-122	44	3	+	+	ADV
ijassa-122	44	4	wtjδt	wtjδt	ADV
ijassa-122	44	5	,	,	PUNCT
ijassa-122	44	6	tj	tj	NOUN
ijassa-122	44	7	)	)	PUNCT
ijassa-122	44	8	=	=	SYM
ijassa-122	44	9	d∑	d∑	PROPN
ijassa-122	44	10	i=1	i=1	X
ijassa-122	44	11	υi(b)xi	υi(b)xi	NOUN
ijassa-122	44	12	,	,	PUNCT
ijassa-122	44	13	t	t	PROPN
ijassa-122	44	14	+	+	CCONJ
ijassa-122	44	15	wtj	wtj	PROPN
ijassa-122	44	16	θ(b	θ(b	NOUN
ijassa-122	44	17	)	)	PUNCT
ijassa-122	44	18	ϕ(b	ϕ(b	PROPN
ijassa-122	44	19	)	)	PUNCT
ijassa-122	44	20	δt	δt	PROPN
ijassa-122	44	21	,	,	PUNCT
ijassa-122	44	22	tj	tj	PROPN
ijassa-122	44	23	+	+	CCONJ
ijassa-122	44	24	θ(b	θ(b	NOUN
ijassa-122	44	25	)	)	PUNCT
ijassa-122	44	26	ϕ(b	ϕ(b	PROPN
ijassa-122	44	27	)	)	PUNCT
ijassa-122	45	1	εt	εt	PROPN
ijassa-122	45	2	.	.	NOUN
ijassa-122	45	3	3	3	NUM
ijassa-122	45	4	identification	identification	NOUN
ijassa-122	45	5	of	of	ADP
ijassa-122	45	6	armax	armax	NOUN
ijassa-122	45	7	models	model	NOUN
ijassa-122	45	8	suppose	suppose	VERB
ijassa-122	45	9	that	that	SCONJ
ijassa-122	45	10	the	the	DET
ijassa-122	45	11	armax	armax	ADJ
ijassa-122	45	12	model	model	NOUN
ijassa-122	45	13	of	of	ADP
ijassa-122	45	14	only	only	ADV
ijassa-122	45	15	one	one	NUM
ijassa-122	45	16	input	input	NOUN
ijassa-122	45	17	process	process	NOUN
ijassa-122	45	18	is	be	AUX
ijassa-122	45	19	as	as	SCONJ
ijassa-122	45	20	follows	follow	VERB
ijassa-122	45	21	:	:	PUNCT
ijassa-122	45	22	zt	zt	PROPN
ijassa-122	45	23	=	=	SYM
ijassa-122	45	24	δ−1(b)ω(b)xt−b	δ−1(b)ω(b)xt−b	PROPN
ijassa-122	45	25	+	+	CCONJ
ijassa-122	45	26	nt	not	PART
ijassa-122	45	27	=	=	NOUN
ijassa-122	45	28	υ(b)xt	υ(b)xt	PROPN
ijassa-122	46	1	+	+	CCONJ
ijassa-122	46	2	nt	not	PART
ijassa-122	46	3	,	,	PUNCT
ijassa-122	46	4	(	(	PUNCT
ijassa-122	46	5	1	1	X
ijassa-122	46	6	)	)	PUNCT
ijassa-122	46	7	where	where	SCONJ
ijassa-122	46	8	δ(b	δ(b	VERB
ijassa-122	46	9	)	)	PUNCT
ijassa-122	46	10	=	=	SYM
ijassa-122	46	11	1−	1−	NUM
ijassa-122	46	12	δ1(b)−	δ1(b)−	X
ijassa-122	46	13	·	·	PUNCT
ijassa-122	46	14	·	·	PUNCT
ijassa-122	46	15	·	·	PUNCT
ijassa-122	47	1	−	−	PRON
ijassa-122	47	2	δr1b	δr1b	PROPN
ijassa-122	47	3	r1	r1	PROPN
ijassa-122	47	4	,	,	PUNCT
ijassa-122	47	5	ω(b	ω(b	NOUN
ijassa-122	47	6	)	)	PUNCT
ijassa-122	47	7	=	=	PUNCT
ijassa-122	47	8	ω0	ω0	ADV
ijassa-122	47	9	−	−	NOUN
ijassa-122	47	10	ω1(b)−	ω1(b)−	SYM
ijassa-122	47	11	·	·	PUNCT
ijassa-122	47	12	·	·	PUNCT
ijassa-122	47	13	·	·	PUNCT
ijassa-122	48	1	−	−	PUNCT
ijassa-122	49	1	ωr2b	ωr2b	NUM
ijassa-122	49	2	r2	r2	NOUN
ijassa-122	49	3	and	and	CCONJ
ijassa-122	49	4	υ(b	υ(b	NOUN
ijassa-122	49	5	)	)	PUNCT
ijassa-122	50	1	=	=	VERB
ijassa-122	50	2	δ−1(b)ω(b)bb	δ−1(b)ω(b)bb	NOUN
ijassa-122	50	3	suppose	suppose	VERB
ijassa-122	50	4	the	the	DET
ijassa-122	50	5	input	input	NOUN
ijassa-122	50	6	process	process	NOUN
ijassa-122	50	7	xt	xt	PROPN
ijassa-122	50	8	is	be	AUX
ijassa-122	50	9	stationary	stationary	ADJ
ijassa-122	50	10	and	and	CCONJ
ijassa-122	50	11	is	be	AUX
ijassa-122	50	12	able	able	ADJ
ijassa-122	50	13	to	to	PART
ijassa-122	50	14	be	be	AUX
ijassa-122	50	15	represented	represent	VERB
ijassa-122	50	16	by	by	ADP
ijassa-122	50	17	some	some	DET
ijassa-122	50	18	member	member	NOUN
ijassa-122	50	19	of	of	ADP
ijassa-122	50	20	the	the	DET
ijassa-122	50	21	general	general	ADJ
ijassa-122	50	22	linear	linear	ADJ
ijassa-122	50	23	class	class	NOUN
ijassa-122	50	24	of	of	ADP
ijassa-122	50	25	autoregressive	autoregressive	ADJ
ijassa-122	50	26	-	-	PUNCT
ijassa-122	50	27	moving	move	VERB
ijassa-122	50	28	average	average	ADJ
ijassa-122	50	29	models	model	NOUN
ijassa-122	50	30	.	.	PUNCT
ijassa-122	51	1	given	give	VERB
ijassa-122	51	2	a	a	DET
ijassa-122	51	3	set	set	NOUN
ijassa-122	51	4	of	of	ADP
ijassa-122	51	5	data	datum	NOUN
ijassa-122	51	6	,	,	PUNCT
ijassa-122	51	7	similar	similar	ADJ
ijassa-122	51	8	to	to	ADP
ijassa-122	51	9	box	box	PROPN
ijassa-122	51	10	et	et	NOUN
ijassa-122	51	11	al.[6	al.[6	PROPN
ijassa-122	51	12	]	]	PUNCT
ijassa-122	51	13	,	,	PUNCT
ijassa-122	51	14	then	then	ADV
ijassa-122	51	15	we	we	PRON
ijassa-122	51	16	can	can	AUX
ijassa-122	51	17	carry	carry	VERB
ijassa-122	51	18	out	out	ADP
ijassa-122	51	19	our	our	PRON
ijassa-122	51	20	usual	usual	ADJ
ijassa-122	51	21	identification	identification	NOUN
ijassa-122	51	22	and	and	CCONJ
ijassa-122	51	23	estimation	estimation	NOUN
ijassa-122	51	24	methods	method	NOUN
ijassa-122	51	25	to	to	PART
ijassa-122	51	26	obtain	obtain	VERB
ijassa-122	51	27	a	a	DET
ijassa-122	51	28	model	model	NOUN
ijassa-122	51	29	for	for	ADP
ijassa-122	51	30	the	the	DET
ijassa-122	51	31	xt	xt	PROPN
ijassa-122	51	32	process	process	NOUN
ijassa-122	51	33	ϕ(b)θ−1(b)xt	ϕ(b)θ−1(b)xt	PROPN
ijassa-122	51	34	=	=	PUNCT
ijassa-122	51	35	αt	αt	NOUN
ijassa-122	51	36	which	which	PRON
ijassa-122	51	37	,	,	PUNCT
ijassa-122	51	38	to	to	ADP
ijassa-122	51	39	a	a	DET
ijassa-122	51	40	close	close	ADJ
ijassa-122	51	41	approximation	approximation	NOUN
ijassa-122	51	42	,	,	PUNCT
ijassa-122	51	43	transforms	transform	VERB
ijassa-122	51	44	the	the	DET
ijassa-122	51	45	correlated	correlate	VERB
ijassa-122	51	46	input	input	NOUN
ijassa-122	51	47	series	series	PROPN
ijassa-122	51	48	xt	xt	PROPN
ijassa-122	51	49	to	to	ADP
ijassa-122	51	50	the	the	DET
ijassa-122	51	51	uncorrelated	uncorrelated	ADJ
ijassa-122	51	52	white	white	PROPN
ijassa-122	51	53	nose	nose	PROPN
ijassa-122	51	54	series	series	PROPN
ijassa-122	51	55	αt	αt	PROPN
ijassa-122	51	56	.	.	PROPN
ijassa-122	52	1	at	at	ADP
ijassa-122	52	2	the	the	DET
ijassa-122	52	3	same	same	ADJ
ijassa-122	52	4	time	time	NOUN
ijassa-122	52	5	,	,	PUNCT
ijassa-122	52	6	we	we	PRON
ijassa-122	52	7	can	can	AUX
ijassa-122	52	8	obtain	obtain	VERB
ijassa-122	52	9	an	an	DET
ijassa-122	52	10	estimate	estimate	NOUN
ijassa-122	52	11	s2α	s2α	ADP
ijassa-122	52	12	of	of	ADP
ijassa-122	52	13	σ2α	σ2α	VERB
ijassa-122	52	14	from	from	ADP
ijassa-122	52	15	the	the	DET
ijassa-122	52	16	sum	sum	NOUN
ijassa-122	52	17	of	of	ADP
ijassa-122	52	18	squares	square	NOUN
ijassa-122	52	19	of	of	ADP
ijassa-122	52	20	the	the	DET
ijassa-122	52	21	α̂′s	α̂′s	PROPN
ijassa-122	52	22	.	.	PUNCT
ijassa-122	53	1	if	if	SCONJ
ijassa-122	53	2	we	we	PRON
ijassa-122	53	3	now	now	ADV
ijassa-122	53	4	apply	apply	VERB
ijassa-122	53	5	this	this	DET
ijassa-122	53	6	same	same	ADJ
ijassa-122	53	7	transformation	transformation	NOUN
ijassa-122	53	8	to	to	ADP
ijassa-122	53	9	zt	zt	PROPN
ijassa-122	53	10	to	to	PART
ijassa-122	53	11	obtain	obtain	VERB
ijassa-122	53	12	βt	βt	NOUN
ijassa-122	53	13	=	=	SYM
ijassa-122	53	14	ϕ(b)θ−1(b)zt	ϕ(b)θ−1(b)zt	PROPN
ijassa-122	53	15	,	,	PUNCT
ijassa-122	53	16	then	then	ADV
ijassa-122	53	17	the	the	DET
ijassa-122	53	18	model(1	model(1	NOUN
ijassa-122	53	19	)	)	PUNCT
ijassa-122	53	20	may	may	AUX
ijassa-122	53	21	be	be	AUX
ijassa-122	53	22	written	write	VERB
ijassa-122	53	23	βt	βt	ADP
ijassa-122	53	24	=	=	SYM
ijassa-122	53	25	υ(b)αt	υ(b)αt	PROPN
ijassa-122	53	26	+	+	CCONJ
ijassa-122	53	27	εt	εt	PROPN
ijassa-122	53	28	,	,	PUNCT
ijassa-122	53	29	multiplying	multiply	VERB
ijassa-122	53	30	αt−k	αt−k	NOUN
ijassa-122	53	31	on	on	ADP
ijassa-122	53	32	both	both	DET
ijassa-122	53	33	sides	side	NOUN
ijassa-122	53	34	and	and	CCONJ
ijassa-122	53	35	taking	take	VERB
ijassa-122	53	36	expectations	expectation	NOUN
ijassa-122	53	37	,	,	PUNCT
ijassa-122	53	38	we	we	PRON
ijassa-122	53	39	obtain	obtain	VERB
ijassa-122	53	40	γαβ(k	γαβ(k	X
ijassa-122	53	41	)	)	PUNCT
ijassa-122	53	42	=	=	VERB
ijassa-122	53	43	υkσ	υkσ	VERB
ijassa-122	53	44	2	2	NUM
ijassa-122	53	45	α	α	NOUN
ijassa-122	53	46	,	,	PUNCT
ijassa-122	53	47	where	where	SCONJ
ijassa-122	53	48	γαβ(k	γαβ(k	NUM
ijassa-122	53	49	)	)	PUNCT
ijassa-122	53	50	=	=	PRON
ijassa-122	54	1	e[αt−kβt]is	e[αt−kβt]is	PROPN
ijassa-122	54	2	the	the	DET
ijassa-122	54	3	cross	cross	NOUN
ijassa-122	54	4	covariance	covariance	NOUN
ijassa-122	54	5	at	at	ADP
ijassa-122	54	6	lag	lag	NOUN
ijassa-122	54	7	k	k	PROPN
ijassa-122	54	8	between	between	ADP
ijassa-122	54	9	α	α	PROPN
ijassa-122	54	10	and	and	CCONJ
ijassa-122	54	11	β	β	NOUN
ijassa-122	54	12	.	.	PUNCT
ijassa-122	54	13	thus	thus	ADV
ijassa-122	54	14	υk	υk	VERB
ijassa-122	54	15	=	=	PUNCT
ijassa-122	55	1	[	[	X
ijassa-122	55	2	ραβ(k)σβ]/[σα	ραβ(k)σβ]/[σα	NUM
ijassa-122	55	3	]	]	X
ijassa-122	55	4	,	,	PUNCT
ijassa-122	55	5	k	k	X
ijassa-122	55	6	=	=	SYM
ijassa-122	55	7	0	0	NUM
ijassa-122	55	8	,	,	PUNCT
ijassa-122	55	9	1	1	NUM
ijassa-122	55	10	,	,	PUNCT
ijassa-122	55	11	2	2	NUM
ijassa-122	55	12	·	·	PUNCT
ijassa-122	55	13	·	·	PUNCT
ijassa-122	55	14	·	·	PUNCT
ijassa-122	55	15	.	.	PUNCT
ijassa-122	56	1	hence	hence	ADV
ijassa-122	56	2	,	,	PUNCT
ijassa-122	56	3	after	after	ADP
ijassa-122	56	4	“	"	PUNCT
ijassa-122	56	5	prewhitening	prewhitene	VERB
ijassa-122	56	6	”	"	PUNCT
ijassa-122	56	7	the	the	DET
ijassa-122	56	8	input	input	NOUN
ijassa-122	56	9	,	,	PUNCT
ijassa-122	56	10	the	the	DET
ijassa-122	56	11	cross	cross	NOUN
ijassa-122	56	12	correlation	correlation	NOUN
ijassa-122	56	13	function	function	NOUN
ijassa-122	56	14	between	between	ADP
ijassa-122	56	15	the	the	DET
ijassa-122	56	16	prewhitened	prewhitene	VERB
ijassa-122	56	17	input	input	NOUN
ijassa-122	56	18	and	and	CCONJ
ijassa-122	56	19	correspondingly	correspondingly	ADV
ijassa-122	56	20	transformed	transform	VERB
ijassa-122	56	21	output	output	NOUN
ijassa-122	56	22	is	be	AUX
ijassa-122	56	23	directly	directly	ADV
ijassa-122	56	24	proportional	proportional	ADJ
ijassa-122	56	25	to	to	ADP
ijassa-122	56	26	the	the	DET
ijassa-122	56	27	response	response	NOUN
ijassa-122	56	28	function	function	NOUN
ijassa-122	56	29	.	.	PUNCT
ijassa-122	57	1	in	in	ADP
ijassa-122	57	2	practice	practice	NOUN
ijassa-122	57	3	,	,	PUNCT
ijassa-122	57	4	we	we	PRON
ijassa-122	57	5	do	do	AUX
ijassa-122	57	6	not	not	PART
ijassa-122	57	7	know	know	VERB
ijassa-122	57	8	the	the	DET
ijassa-122	57	9	theoretical	theoretical	ADJ
ijassa-122	57	10	function	function	NOUN
ijassa-122	57	11	ραβ(k	ραβ(k	NUM
ijassa-122	57	12	)	)	PUNCT
ijassa-122	57	13	,	,	PUNCT
ijassa-122	57	14	so	so	SCONJ
ijassa-122	57	15	we	we	PRON
ijassa-122	57	16	must	must	AUX
ijassa-122	57	17	substitute	substitute	VERB
ijassa-122	57	18	estimates	estimate	NOUN
ijassa-122	57	19	in	in	ADP
ijassa-122	57	20	υk	υk	NOUN
ijassa-122	57	21	to	to	PART
ijassa-122	57	22	give	give	VERB
ijassa-122	57	23	υ̂k	υ̂k	NUM
ijassa-122	57	24	=	=	PUNCT
ijassa-122	58	1	[	[	X
ijassa-122	58	2	rαβ(k)sβ]/[sα	rαβ(k)sβ]/[sα	X
ijassa-122	58	3	]	]	PUNCT
ijassa-122	58	4	,	,	PUNCT
ijassa-122	58	5	k	k	PROPN
ijassa-122	58	6	=	=	SYM
ijassa-122	58	7	0	0	NUM
ijassa-122	58	8	,	,	PUNCT
ijassa-122	58	9	1	1	NUM
ijassa-122	58	10	,	,	PUNCT
ijassa-122	58	11	2	2	NUM
ijassa-122	58	12	...	...	PUNCT
ijassa-122	58	13	where	where	SCONJ
ijassa-122	58	14	rαβ(k	rαβ(k	X
ijassa-122	58	15	)	)	PUNCT
ijassa-122	58	16	=	=	SYM
ijassa-122	58	17	cαβ(k)/[sα	cαβ(k)/[sα	X
ijassa-122	58	18	/	/	SYM
ijassa-122	58	19	sβ	sβ	NOUN
ijassa-122	58	20	]	]	PUNCT
ijassa-122	58	21	,	,	PUNCT
ijassa-122	58	22	cαβ(k	cαβ(k	PROPN
ijassa-122	58	23	)	)	PUNCT
ijassa-122	58	24	=	=	SYM
ijassa-122	59	1	1	1	NUM
ijassa-122	59	2	n	n	NUM
ijassa-122	59	3	n−k∑	n−k∑	NOUN
ijassa-122	59	4	n=1	n=1	PROPN
ijassa-122	59	5	(	(	PUNCT
ijassa-122	59	6	αt	αt	NOUN
ijassa-122	59	7	−	−	PROPN
ijassa-122	59	8	ᾱ)(βt+k	ᾱ)(βt+k	PROPN
ijassa-122	59	9	−	−	PROPN
ijassa-122	59	10	β̄	β̄	NOUN
ijassa-122	59	11	)	)	PUNCT
ijassa-122	59	12	,	,	PUNCT
ijassa-122	59	13	sα	sα	ADJ
ijassa-122	59	14	=	=	NOUN
ijassa-122	59	15	√	√	PROPN
ijassa-122	59	16	cαα(0	cαα(0	NOUN
ijassa-122	59	17	)	)	PUNCT
ijassa-122	59	18	,	,	PUNCT
ijassa-122	59	19	sβ	sβ	NOUN
ijassa-122	59	20	=	=	PUNCT
ijassa-122	59	21	√	√	NOUN
ijassa-122	59	22	cββ(0	cββ(0	NOUN
ijassa-122	59	23	)	)	PUNCT
ijassa-122	59	24	,	,	PUNCT
ijassa-122	59	25	k	k	PROPN
ijassa-122	59	26	=	=	SYM
ijassa-122	59	27	0	0	NUM
ijassa-122	59	28	,	,	PUNCT
ijassa-122	59	29	1	1	NUM
ijassa-122	59	30	,	,	PUNCT
ijassa-122	59	31	2	2	NUM
ijassa-122	59	32	,	,	PUNCT
ijassa-122	59	33	...	...	PUNCT
ijassa-122	59	34	402	402	NUM
ijassa-122	59	35	ping	ping	NOUN
ijassa-122	59	36	chen	chen	PROPN
ijassa-122	59	37	:	:	PUNCT
ijassa-122	59	38	the	the	DET
ijassa-122	59	39	identification	identification	NOUN
ijassa-122	59	40	of	of	ADP
ijassa-122	59	41	outliers	outlier	NOUN
ijassa-122	59	42	in	in	ADP
ijassa-122	59	43	armax	armax	NOUN
ijassa-122	59	44	models	model	NOUN
ijassa-122	59	45	via	via	ADP
ijassa-122	59	46	genetic	genetic	ADJ
ijassa-122	59	47	algorithm	algorithm	NOUN
ijassa-122	59	48	the	the	DET
ijassa-122	59	49	preliminary	preliminary	ADJ
ijassa-122	59	50	estimates	estimate	NOUN
ijassa-122	59	51	υ̂k	υ̂k	NUM
ijassa-122	59	52	can	can	AUX
ijassa-122	59	53	provide	provide	VERB
ijassa-122	59	54	a	a	DET
ijassa-122	59	55	rough	rough	ADJ
ijassa-122	59	56	basis	basis	NOUN
ijassa-122	59	57	for	for	ADP
ijassa-122	59	58	selecting	select	VERB
ijassa-122	59	59	suitable	suitable	ADJ
ijassa-122	59	60	transfer	transfer	NOUN
ijassa-122	59	61	function	function	NOUN
ijassa-122	59	62	model	model	NOUN
ijassa-122	59	63	.	.	PUNCT
ijassa-122	60	1	first	first	ADV
ijassa-122	60	2	,	,	PUNCT
ijassa-122	60	3	we	we	PRON
ijassa-122	60	4	may	may	AUX
ijassa-122	60	5	use	use	VERB
ijassa-122	60	6	the	the	DET
ijassa-122	60	7	estimates	estimate	NOUN
ijassa-122	60	8	υ̂k	υ̂k	PUNCT
ijassa-122	60	9	so	so	ADV
ijassa-122	60	10	obtained	obtain	VERB
ijassa-122	60	11	to	to	PART
ijassa-122	60	12	make	make	VERB
ijassa-122	60	13	guesses	guess	NOUN
ijassa-122	60	14	of	of	ADP
ijassa-122	60	15	the	the	DET
ijassa-122	60	16	order	order	NOUN
ijassa-122	60	17	r1	r1	NOUN
ijassa-122	60	18	and	and	CCONJ
ijassa-122	60	19	r2	r2	PROPN
ijassa-122	60	20	of	of	ADP
ijassa-122	60	21	δ(b	δ(b	PROPN
ijassa-122	60	22	)	)	PUNCT
ijassa-122	60	23	and	and	CCONJ
ijassa-122	60	24	ω(b	ω(b	NOUN
ijassa-122	60	25	)	)	PUNCT
ijassa-122	60	26	,	,	PUNCT
ijassa-122	60	27	and	and	CCONJ
ijassa-122	60	28	of	of	ADP
ijassa-122	60	29	the	the	DET
ijassa-122	60	30	delay	delay	PROPN
ijassa-122	60	31	parameter	parameter	PROPN
ijassa-122	60	32	b.	b.	PROPN
ijassa-122	60	33	second	second	PROPN
ijassa-122	60	34	,	,	PUNCT
ijassa-122	60	35	we	we	PRON
ijassa-122	60	36	do	do	AUX
ijassa-122	60	37	not	not	PART
ijassa-122	60	38	consider	consider	VERB
ijassa-122	60	39	the	the	DET
ijassa-122	60	40	noise	noise	NOUN
ijassa-122	60	41	nt	not	PART
ijassa-122	60	42	now	now	ADV
ijassa-122	60	43	,	,	PUNCT
ijassa-122	60	44	substituting	substitute	VERB
ijassa-122	60	45	zt	zt	PROPN
ijassa-122	60	46	=	=	SYM
ijassa-122	60	47	υ̂(b)xt	υ̂(b)xt	PROPN
ijassa-122	60	48	in	in	ADP
ijassa-122	60	49	the	the	DET
ijassa-122	60	50	equation	equation	NOUN
ijassa-122	60	51	δ(b)zt	δ(b)zt	NUM
ijassa-122	60	52	=	=	SYM
ijassa-122	60	53	ω(b)bbxt	ω(b)bbxt	NUM
ijassa-122	60	54	,	,	PUNCT
ijassa-122	60	55	based	base	VERB
ijassa-122	60	56	on	on	ADP
ijassa-122	60	57	equating	equate	VERB
ijassa-122	60	58	coefficients	coefficient	NOUN
ijassa-122	60	59	of	of	ADP
ijassa-122	60	60	b	b	NOUN
ijassa-122	60	61	,	,	PUNCT
ijassa-122	60	62	to	to	PART
ijassa-122	60	63	obtain	obtain	VERB
ijassa-122	60	64	initial	initial	ADJ
ijassa-122	60	65	estimates	estimate	NOUN
ijassa-122	60	66	of	of	ADP
ijassa-122	60	67	the	the	DET
ijassa-122	60	68	parameters	parameter	NOUN
ijassa-122	60	69	δ(b	δ(b	VERB
ijassa-122	60	70	)	)	PUNCT
ijassa-122	60	71	and	and	CCONJ
ijassa-122	60	72	ω(b	ω(b	NOUN
ijassa-122	60	73	)	)	PUNCT
ijassa-122	60	74	.	.	PUNCT
ijassa-122	61	1	4	4	NUM
ijassa-122	61	2	the	the	DET
ijassa-122	61	3	identification	identification	NOUN
ijassa-122	61	4	of	of	ADP
ijassa-122	61	5	outliers	outlier	NOUN
ijassa-122	61	6	via	via	ADP
ijassa-122	61	7	genetic	genetic	ADJ
ijassa-122	61	8	algorithm	algorithm	NOUN
ijassa-122	61	9	we	we	PRON
ijassa-122	61	10	let	let	VERB
ijassa-122	61	11	{	{	PUNCT
ijassa-122	61	12	yt	yt	PROPN
ijassa-122	61	13	,	,	PUNCT
ijassa-122	61	14	t	t	NOUN
ijassa-122	61	15	=	=	SYM
ijassa-122	61	16	0	0	NUM
ijassa-122	61	17	,	,	PUNCT
ijassa-122	61	18	1	1	NUM
ijassa-122	61	19	,	,	PUNCT
ijassa-122	61	20	2	2	NUM
ijassa-122	61	21	,	,	PUNCT
ijassa-122	61	22	·	·	PUNCT
ijassa-122	61	23	·	·	PUNCT
ijassa-122	61	24	·	·	PUNCT
ijassa-122	61	25	}	}	PUNCT
ijassa-122	61	26	be	be	AUX
ijassa-122	61	27	a	a	DET
ijassa-122	61	28	zero	zero	NUM
ijassa-122	61	29	mean	mean	NOUN
ijassa-122	61	30	and	and	CCONJ
ijassa-122	61	31	stationary	stationary	ADJ
ijassa-122	61	32	time	time	NOUN
ijassa-122	61	33	series	series	PROPN
ijassa-122	61	34	:	:	PUNCT
ijassa-122	61	35	yt	yt	PROPN
ijassa-122	61	36	=	=	PUNCT
ijassa-122	61	37	∞∑	∞∑	NUM
ijassa-122	61	38	j=0	j=0	PROPN
ijassa-122	61	39	ψjαt−j	ψjαt−j	PROPN
ijassa-122	61	40	,	,	PUNCT
ijassa-122	61	41	where	where	SCONJ
ijassa-122	61	42	{	{	PUNCT
ijassa-122	61	43	αt	αt	NOUN
ijassa-122	61	44	}	}	PUNCT
ijassa-122	61	45	is	be	AUX
ijassa-122	61	46	gaussian	gaussian	ADJ
ijassa-122	61	47	zero	zero	NUM
ijassa-122	61	48	mean	mean	VERB
ijassa-122	61	49	white	white	ADJ
ijassa-122	61	50	noise	noise	NOUN
ijassa-122	61	51	and	and	CCONJ
ijassa-122	61	52	v	v	NOUN
ijassa-122	61	53	ar(αt	ar(αt	NOUN
ijassa-122	61	54	)	)	PUNCT
ijassa-122	61	55	=	=	SYM
ijassa-122	61	56	σ2	σ2	NOUN
ijassa-122	61	57	,	,	PUNCT
ijassa-122	61	58	{	{	PUNCT
ijassa-122	61	59	ψj	ψj	ADV
ijassa-122	61	60	,	,	PUNCT
ijassa-122	61	61	j	j	PROPN
ijassa-122	61	62	=	=	SYM
ijassa-122	61	63	0	0	NUM
ijassa-122	61	64	,	,	PUNCT
ijassa-122	61	65	1	1	NUM
ijassa-122	61	66	,	,	PUNCT
ijassa-122	61	67	2	2	NUM
ijassa-122	61	68	,	,	PUNCT
ijassa-122	61	69	·	·	PUNCT
ijassa-122	61	70	·	·	PUNCT
ijassa-122	61	71	·	·	PUNCT
ijassa-122	61	72	}	}	PUNCT
ijassa-122	61	73	form	form	VERB
ijassa-122	61	74	a	a	DET
ijassa-122	61	75	absolutely	absolutely	ADV
ijassa-122	61	76	summable	summable	ADJ
ijassa-122	61	77	sequence	sequence	NOUN
ijassa-122	61	78	.	.	PUNCT
ijassa-122	62	1	let	let	VERB
ijassa-122	62	2	γih	γih	NOUN
ijassa-122	62	3	denote	denote	VERB
ijassa-122	62	4	the	the	DET
ijassa-122	62	5	inverse	inverse	NOUN
ijassa-122	62	6	autocovariance	autocovariance	NOUN
ijassa-122	62	7	function	function	NOUN
ijassa-122	62	8	of	of	ADP
ijassa-122	62	9	the	the	DET
ijassa-122	62	10	process	process	NOUN
ijassa-122	62	11	,	,	PUNCT
ijassa-122	62	12	for	for	ADP
ijassa-122	62	13	integer	integer	PROPN
ijassa-122	62	14	h.	h.	PROPN
ijassa-122	62	15	also	also	ADV
ijassa-122	62	16	let	let	VERB
ijassa-122	62	17	ρih	ρih	NOUN
ijassa-122	62	18	=	=	SYM
ijassa-122	62	19	γih	γih	PROPN
ijassa-122	62	20	/	/	SYM
ijassa-122	62	21	γi0	γi0	PROPN
ijassa-122	63	1	denote	denote	VERB
ijassa-122	63	2	the	the	DET
ijassa-122	63	3	inverse	inverse	NOUN
ijassa-122	63	4	autocorrelations	autocorrelation	NOUN
ijassa-122	63	5	.	.	PUNCT
ijassa-122	64	1	we	we	PRON
ijassa-122	64	2	have	have	VERB
ijassa-122	64	3	γik	γik	NOUN
ijassa-122	65	1	+	+	CCONJ
ijassa-122	65	2	ψ1γik−1	ψ1γik−1	PROPN
ijassa-122	65	3	+	+	CCONJ
ijassa-122	65	4	ψ2γik−2	ψ2γik−2	NOUN
ijassa-122	65	5	+	+	CCONJ
ijassa-122	65	6	·	·	PUNCT
ijassa-122	65	7	·	·	PUNCT
ijassa-122	65	8	·	·	PUNCT
ijassa-122	66	1	=	=	SYM
ijassa-122	66	2	0	0	NUM
ijassa-122	66	3	,	,	PUNCT
ijassa-122	66	4	k	k	PROPN
ijassa-122	66	5	>	>	X
ijassa-122	66	6	0	0	PUNCT
ijassa-122	66	7	(	(	PUNCT
ijassa-122	66	8	2	2	NUM
ijassa-122	66	9	)	)	PUNCT
ijassa-122	66	10	when	when	SCONJ
ijassa-122	66	11	outliers	outlier	NOUN
ijassa-122	66	12	are	be	AUX
ijassa-122	66	13	present	present	ADJ
ijassa-122	66	14	,	,	PUNCT
ijassa-122	66	15	{	{	PUNCT
ijassa-122	66	16	yt	yt	PROPN
ijassa-122	66	17	,	,	PUNCT
ijassa-122	66	18	t	t	NOUN
ijassa-122	66	19	=	=	SYM
ijassa-122	66	20	0	0	NUM
ijassa-122	66	21	,	,	PUNCT
ijassa-122	66	22	1	1	NUM
ijassa-122	66	23	,	,	PUNCT
ijassa-122	66	24	2	2	NUM
ijassa-122	66	25	,	,	PUNCT
ijassa-122	66	26	·	·	PUNCT
ijassa-122	66	27	·	·	PUNCT
ijassa-122	66	28	·	·	PUNCT
ijassa-122	66	29	}	}	PUNCT
ijassa-122	66	30	is	be	AUX
ijassa-122	66	31	unobservable	unobservable	ADJ
ijassa-122	66	32	.	.	PUNCT
ijassa-122	67	1	instead	instead	ADV
ijassa-122	67	2	the	the	DET
ijassa-122	67	3	time	time	NOUN
ijassa-122	67	4	series	series	PROPN
ijassa-122	67	5	{	{	PUNCT
ijassa-122	67	6	zt	zt	PROPN
ijassa-122	67	7	,	,	PUNCT
ijassa-122	67	8	t	t	PROPN
ijassa-122	67	9	=	=	SYM
ijassa-122	67	10	0	0	NUM
ijassa-122	67	11	,	,	PUNCT
ijassa-122	67	12	1	1	NUM
ijassa-122	67	13	,	,	PUNCT
ijassa-122	67	14	2	2	NUM
ijassa-122	67	15	,	,	PUNCT
ijassa-122	67	16	·	·	PUNCT
ijassa-122	67	17	·	·	PUNCT
ijassa-122	67	18	·	·	PUNCT
ijassa-122	67	19	}	}	PUNCT
ijassa-122	67	20	is	be	AUX
ijassa-122	67	21	observed	observe	VERB
ijassa-122	67	22	which	which	PRON
ijassa-122	67	23	follows	follow	VERB
ijassa-122	67	24	the	the	DET
ijassa-122	67	25	model	model	NOUN
ijassa-122	67	26	:	:	PUNCT
ijassa-122	67	27	zt	zt	PROPN
ijassa-122	67	28	=	=	SYM
ijassa-122	67	29	yt	yt	PROPN
ijassa-122	67	30	+	+	CCONJ
ijassa-122	67	31	dt	dt	PROPN
ijassa-122	67	32	,	,	PUNCT
ijassa-122	67	33	where	where	SCONJ
ijassa-122	67	34	dt	dt	PROPN
ijassa-122	67	35	is	be	AUX
ijassa-122	67	36	a	a	DET
ijassa-122	67	37	deterministic	deterministic	ADJ
ijassa-122	67	38	perturbation	perturbation	NOUN
ijassa-122	67	39	.	.	PUNCT
ijassa-122	68	1	let	let	VERB
ijassa-122	68	2	ψj	ψj	ADV
ijassa-122	68	3	=	=	NOUN
ijassa-122	68	4	0	0	PUNCT
ijassa-122	69	1	if	if	SCONJ
ijassa-122	69	2	j	j	PROPN
ijassa-122	69	3	<	<	X
ijassa-122	69	4	0	0	PROPN
ijassa-122	69	5	,	,	PUNCT
ijassa-122	69	6	then	then	ADV
ijassa-122	69	7	we	we	PRON
ijassa-122	69	8	have	have	VERB
ijassa-122	69	9	dt	dt	NOUN
ijassa-122	69	10	=	=	SYM
ijassa-122	69	11	ψt−t0w0	ψt−t0w0	PROPN
ijassa-122	69	12	,	,	PUNCT
ijassa-122	69	13	if	if	SCONJ
ijassa-122	69	14	time	time	NOUN
ijassa-122	69	15	is	be	AUX
ijassa-122	69	16	io	io	NOUN
ijassa-122	69	17	;	;	PUNCT
ijassa-122	69	18	or	or	CCONJ
ijassa-122	69	19	dt	dt	NOUN
ijassa-122	69	20	=	=	SYM
ijassa-122	69	21	w0δt	w0δt	X
ijassa-122	69	22	,	,	PUNCT
ijassa-122	69	23	t0	t0	PROPN
ijassa-122	69	24	,	,	PUNCT
ijassa-122	69	25	if	if	SCONJ
ijassa-122	69	26	time	time	NOUN
ijassa-122	69	27	t0	t0	PROPN
ijassa-122	69	28	is	be	AUX
ijassa-122	69	29	ao	ao	PROPN
ijassa-122	69	30	,	,	PUNCT
ijassa-122	69	31	where	where	SCONJ
ijassa-122	69	32	w0	w0	PROPN
ijassa-122	69	33	denotes	denote	VERB
ijassa-122	69	34	the	the	DET
ijassa-122	69	35	outlier	outlier	NOUN
ijassa-122	69	36	’s	’s	PART
ijassa-122	69	37	magnitude	magnitude	NOUN
ijassa-122	69	38	at	at	ADP
ijassa-122	69	39	t	t	PROPN
ijassa-122	69	40	=	=	SYM
ijassa-122	69	41	t0	t0	PROPN
ijassa-122	69	42	.	.	PUNCT
ijassa-122	70	1	in	in	ADP
ijassa-122	70	2	the	the	DET
ijassa-122	70	3	present	present	ADJ
ijassa-122	70	4	setting	setting	NOUN
ijassa-122	70	5	,	,	PUNCT
ijassa-122	70	6	a	a	DET
ijassa-122	70	7	chromosome	chromosome	NOUN
ijassa-122	70	8	ξ	ξ	PROPN
ijassa-122	70	9	is	be	AUX
ijassa-122	70	10	a	a	DET
ijassa-122	70	11	string	string	NOUN
ijassa-122	70	12	of	of	ADP
ijassa-122	70	13	characters	character	NOUN
ijassa-122	70	14	of	of	ADP
ijassa-122	70	15	assigned	assign	VERB
ijassa-122	70	16	length	length	NOUN
ijassa-122	70	17	n	n	PROPN
ijassa-122	70	18	that	that	PRON
ijassa-122	70	19	can	can	AUX
ijassa-122	70	20	be	be	AUX
ijassa-122	70	21	evaluated	evaluate	VERB
ijassa-122	70	22	in	in	ADP
ijassa-122	70	23	terms	term	NOUN
ijassa-122	70	24	of	of	ADP
ijassa-122	70	25	the	the	DET
ijassa-122	70	26	ff	ff	NOUN
ijassa-122	70	27	,	,	PUNCT
ijassa-122	70	28	where	where	SCONJ
ijassa-122	70	29	n	n	PRON
ijassa-122	70	30	is	be	AUX
ijassa-122	70	31	the	the	DET
ijassa-122	70	32	number	number	NOUN
ijassa-122	70	33	of	of	ADP
ijassa-122	70	34	observations	observation	NOUN
ijassa-122	70	35	of	of	ADP
ijassa-122	70	36	the	the	DET
ijassa-122	70	37	time	time	NOUN
ijassa-122	70	38	series	series	NOUN
ijassa-122	70	39	,	,	PUNCT
ijassa-122	70	40	as	as	SCONJ
ijassa-122	70	41	each	each	DET
ijassa-122	70	42	locus	locus	NOUN
ijassa-122	70	43	ξj	ξj	NOUN
ijassa-122	70	44	is	be	AUX
ijassa-122	70	45	corresponding	correspond	VERB
ijassa-122	70	46	to	to	ADP
ijassa-122	70	47	an	an	DET
ijassa-122	70	48	observation	observation	NOUN
ijassa-122	70	49	zj	zj	INTJ
ijassa-122	70	50	where	where	SCONJ
ijassa-122	70	51	an	an	DET
ijassa-122	70	52	outlier	outlier	NOUN
ijassa-122	70	53	may	may	AUX
ijassa-122	70	54	occur	occur	VERB
ijassa-122	70	55	.	.	PUNCT
ijassa-122	71	1	so	so	ADV
ijassa-122	71	2	,	,	PUNCT
ijassa-122	71	3	ξ	ξ	X
ijassa-122	71	4	=	=	SYM
ijassa-122	71	5	(	(	PUNCT
ijassa-122	71	6	ξ1	ξ1	PROPN
ijassa-122	71	7	,	,	PUNCT
ijassa-122	71	8	ξ2	ξ2	ADJ
ijassa-122	71	9	,	,	PUNCT
ijassa-122	71	10	·	·	PUNCT
ijassa-122	71	11	·	·	PUNCT
ijassa-122	71	12	·	·	PUNCT
ijassa-122	71	13	,	,	PUNCT
ijassa-122	71	14	ξn	ξn	PROPN
ijassa-122	71	15	)	)	PUNCT
ijassa-122	71	16	.	.	PUNCT
ijassa-122	72	1	then	then	ADV
ijassa-122	72	2	a	a	DET
ijassa-122	72	3	gene	gene	NOUN
ijassa-122	72	4	ξj	ξj	NOUN
ijassa-122	72	5	=	=	NOUN
ijassa-122	72	6	0	0	PROPN
ijassa-122	72	7	,	,	PUNCT
ijassa-122	72	8	if	if	SCONJ
ijassa-122	72	9	the	the	DET
ijassa-122	72	10	locus	locus	NOUN
ijassa-122	72	11	is	be	AUX
ijassa-122	72	12	an	an	DET
ijassa-122	72	13	outlier	outlier	NOUN
ijassa-122	72	14	-	-	PUNCT
ijassa-122	72	15	free	free	ADJ
ijassa-122	72	16	time	time	NOUN
ijassa-122	72	17	point	point	NOUN
ijassa-122	72	18	,	,	PUNCT
ijassa-122	72	19	ξj	ξj	NOUN
ijassa-122	72	20	=	=	NOUN
ijassa-122	72	21	1	1	NUM
ijassa-122	72	22	if	if	SCONJ
ijassa-122	72	23	the	the	DET
ijassa-122	72	24	observation	observation	NOUN
ijassa-122	72	25	at	at	ADP
ijassa-122	72	26	this	this	DET
ijassa-122	72	27	time	time	NOUN
ijassa-122	72	28	point	point	NOUN
ijassa-122	72	29	is	be	AUX
ijassa-122	72	30	an	an	DET
ijassa-122	72	31	additive	additive	ADJ
ijassa-122	72	32	outlier	outlier	NOUN
ijassa-122	72	33	,	,	PUNCT
ijassa-122	72	34	and	and	CCONJ
ijassa-122	72	35	ξj	ξj	NOUN
ijassa-122	72	36	=	=	NOUN
ijassa-122	72	37	2	2	NUM
ijassa-122	72	38	if	if	SCONJ
ijassa-122	72	39	it	it	PRON
ijassa-122	72	40	is	be	AUX
ijassa-122	72	41	an	an	DET
ijassa-122	72	42	innovational	innovational	ADJ
ijassa-122	72	43	outlier	outlier	NOUN
ijassa-122	72	44	.	.	PUNCT
ijassa-122	73	1	if	if	SCONJ
ijassa-122	73	2	k	k	PROPN
ijassa-122	73	3	outliers	outlier	NOUN
ijassa-122	73	4	are	be	AUX
ijassa-122	73	5	located	locate	VERB
ijassa-122	73	6	at	at	ADP
ijassa-122	73	7	t1	t1	NOUN
ijassa-122	73	8	,	,	PUNCT
ijassa-122	73	9	t2	t2	NOUN
ijassa-122	73	10	,	,	PUNCT
ijassa-122	73	11	·	·	PUNCT
ijassa-122	73	12	·	·	PUNCT
ijassa-122	73	13	·	·	PUNCT
ijassa-122	73	14	,	,	PUNCT
ijassa-122	73	15	tk	tk	PROPN
ijassa-122	73	16	and	and	CCONJ
ijassa-122	73	17	denoting	denote	VERB
ijassa-122	73	18	by	by	ADP
ijassa-122	73	19	z	z	NOUN
ijassa-122	73	20	=	=	SYM
ijassa-122	73	21	(	(	PUNCT
ijassa-122	73	22	z1	z1	PROPN
ijassa-122	73	23	,	,	PUNCT
ijassa-122	73	24	z2	z2	PROPN
ijassa-122	73	25	,	,	PUNCT
ijassa-122	73	26	·	·	PUNCT
ijassa-122	73	27	·	·	PUNCT
ijassa-122	73	28	·	·	PUNCT
ijassa-122	73	29	,	,	PUNCT
ijassa-122	73	30	zn)′	zn)′	PRON
ijassa-122	73	31	the	the	DET
ijassa-122	73	32	observed	observe	VERB
ijassa-122	73	33	time	time	NOUN
ijassa-122	73	34	series	series	NOUN
ijassa-122	73	35	and	and	CCONJ
ijassa-122	73	36	by	by	ADP
ijassa-122	73	37	y	y	PROPN
ijassa-122	73	38	=	=	SYM
ijassa-122	73	39	(	(	PUNCT
ijassa-122	73	40	y1	y1	PROPN
ijassa-122	73	41	,	,	PUNCT
ijassa-122	73	42	y2	y2	PROPN
ijassa-122	73	43	,	,	PUNCT
ijassa-122	73	44	·	·	PUNCT
ijassa-122	73	45	·	·	PUNCT
ijassa-122	73	46	·	·	PUNCT
ijassa-122	73	47	,	,	PUNCT
ijassa-122	73	48	yn)′	yn)′	PRON
ijassa-122	73	49	the	the	DET
ijassa-122	73	50	unobserved	unobserved	ADJ
ijassa-122	73	51	realization	realization	NOUN
ijassa-122	73	52	of	of	ADP
ijassa-122	73	53	yt	yt	PROPN
ijassa-122	73	54	,	,	PUNCT
ijassa-122	73	55	then	then	ADV
ijassa-122	73	56	z	z	NOUN
ijassa-122	73	57	=	=	PUNCT
ijassa-122	73	58	ψw	ψw	VERB
ijassa-122	74	1	+	+	ADJ
ijassa-122	74	2	y	y	PROPN
ijassa-122	74	3	,	,	PUNCT
ijassa-122	74	4	where	where	SCONJ
ijassa-122	74	5	w	w	NOUN
ijassa-122	74	6	=	=	SYM
ijassa-122	74	7	(	(	PUNCT
ijassa-122	74	8	w1	w1	NOUN
ijassa-122	74	9	,	,	PUNCT
ijassa-122	74	10	w2	w2	NOUN
ijassa-122	74	11	,	,	PUNCT
ijassa-122	74	12	·	·	PUNCT
ijassa-122	74	13	·	·	PUNCT
ijassa-122	74	14	·	·	PUNCT
ijassa-122	74	15	,	,	PUNCT
ijassa-122	74	16	wk	wk	X
ijassa-122	74	17	)	)	PUNCT
ijassa-122	74	18	′	′	NUM
ijassa-122	74	19	is	be	AUX
ijassa-122	74	20	the	the	DET
ijassa-122	74	21	vector	vector	NOUN
ijassa-122	74	22	of	of	ADP
ijassa-122	74	23	the	the	DET
ijassa-122	74	24	outliers	outlier	NOUN
ijassa-122	74	25	’	'	PUNCT
ijassa-122	74	26	magnitudes	magnitude	NOUN
ijassa-122	74	27	at	at	ADP
ijassa-122	74	28	t1	t1	NOUN
ijassa-122	74	29	,	,	PUNCT
ijassa-122	74	30	t2	t2	NOUN
ijassa-122	74	31	,	,	PUNCT
ijassa-122	74	32	·	·	PUNCT
ijassa-122	74	33	·	·	PUNCT
ijassa-122	74	34	·	·	PUNCT
ijassa-122	74	35	,	,	PUNCT
ijassa-122	74	36	tk	tk	PROPN
ijassa-122	74	37	and	and	CCONJ
ijassa-122	74	38	ψ	ψ	PROPN
ijassa-122	74	39	is	be	AUX
ijassa-122	74	40	the	the	DET
ijassa-122	74	41	matrix	matrix	NOUN
ijassa-122	74	42	of	of	ADP
ijassa-122	74	43	the	the	DET
ijassa-122	74	44	n	n	PROPN
ijassa-122	74	45	×	×	NOUN
ijassa-122	74	46	k	k	PROPN
ijassa-122	74	47	elements	element	NOUN
ijassa-122	74	48	ψjh	ψjh	NOUN
ijassa-122	74	49	defined	define	VERB
ijassa-122	74	50	as	as	SCONJ
ijassa-122	74	51	follows	follow	VERB
ijassa-122	74	52	:	:	PUNCT
ijassa-122	74	53	if	if	SCONJ
ijassa-122	74	54	th	th	X
ijassa-122	74	55	is	be	AUX
ijassa-122	74	56	the	the	DET
ijassa-122	74	57	timing	timing	NOUN
ijassa-122	74	58	of	of	ADP
ijassa-122	74	59	an	an	DET
ijassa-122	74	60	io	io	NOUN
ijassa-122	74	61	,	,	PUNCT
ijassa-122	74	62	then	then	ADV
ijassa-122	74	63	ψjh	ψjh	NOUN
ijassa-122	75	1	=	=	PUNCT
ijassa-122	75	2	ψj−th	ψj−th	PROPN
ijassa-122	76	1	if	if	SCONJ
ijassa-122	76	2	j	j	PROPN
ijassa-122	76	3	>	>	X
ijassa-122	76	4	th	th	X
ijassa-122	76	5	and	and	CCONJ
ijassa-122	76	6	0	0	NUM
ijassa-122	76	7	,	,	PUNCT
ijassa-122	76	8	otherwise	otherwise	ADV
ijassa-122	76	9	.	.	PUNCT
ijassa-122	77	1	if	if	SCONJ
ijassa-122	77	2	an	an	DET
ijassa-122	77	3	ao	ao	NOUN
ijassa-122	77	4	is	be	AUX
ijassa-122	77	5	occurring	occur	VERB
ijassa-122	77	6	in	in	ADP
ijassa-122	77	7	th	th	ADP
ijassa-122	77	8	,	,	PUNCT
ijassa-122	77	9	then	then	ADV
ijassa-122	77	10	ψjh	ψjh	NOUN
ijassa-122	77	11	=	=	NOUN
ijassa-122	77	12	1	1	NUM
ijassa-122	77	13	if	if	SCONJ
ijassa-122	77	14	j	j	PROPN
ijassa-122	77	15	=	=	X
ijassa-122	77	16	th	th	PROPN
ijassa-122	77	17	and	and	CCONJ
ijassa-122	77	18	0	0	NUM
ijassa-122	77	19	,	,	PUNCT
ijassa-122	77	20	otherwise	otherwise	ADV
ijassa-122	77	21	.	.	PUNCT
ijassa-122	78	1	the	the	DET
ijassa-122	78	2	idea	idea	NOUN
ijassa-122	78	3	is	be	AUX
ijassa-122	78	4	to	to	PART
ijassa-122	78	5	seek	seek	VERB
ijassa-122	78	6	for	for	ADP
ijassa-122	78	7	the	the	DET
ijassa-122	78	8	matrix	matrix	NOUN
ijassa-122	78	9	ψ	ψ	X
ijassa-122	78	10	which	which	PRON
ijassa-122	78	11	maximizes	maximize	VERB
ijassa-122	78	12	the	the	DET
ijassa-122	78	13	likelihood	likelihood	NOUN
ijassa-122	78	14	function	function	NOUN
ijassa-122	78	15	.	.	PUNCT
ijassa-122	79	1	when	when	SCONJ
ijassa-122	79	2	n	n	PRON
ijassa-122	79	3	is	be	AUX
ijassa-122	79	4	large	large	ADJ
ijassa-122	79	5	,	,	PUNCT
ijassa-122	79	6	γ−1	γ−1	PROPN
ijassa-122	79	7	may	may	AUX
ijassa-122	79	8	be	be	AUX
ijassa-122	79	9	replaced	replace	VERB
ijassa-122	79	10	by	by	ADP
ijassa-122	79	11	the	the	DET
ijassa-122	79	12	matrix	matrix	NOUN
ijassa-122	79	13	of	of	ADP
ijassa-122	79	14	inverse	inverse	NOUN
ijassa-122	79	15	autocovariances	autocovariance	NOUN
ijassa-122	79	16	γi	γi	INTJ
ijassa-122	79	17	.	.	PUNCT
ijassa-122	80	1	then	then	ADV
ijassa-122	80	2	we	we	PRON
ijassa-122	80	3	have	have	VERB
ijassa-122	80	4	the	the	DET
ijassa-122	80	5	likelihood	likelihood	NOUN
ijassa-122	80	6	function	function	NOUN
ijassa-122	80	7	of	of	ADP
ijassa-122	80	8	the	the	DET
ijassa-122	80	9	time	time	NOUN
ijassa-122	80	10	series	series	PROPN
ijassa-122	80	11	z	z	PROPN
ijassa-122	80	12	l	l	NOUN
ijassa-122	81	1	=	=	PUNCT
ijassa-122	81	2	p(z	p(z	NOUN
ijassa-122	81	3	|	|	ADV
ijassa-122	81	4	ξ	ξ	PROPN
ijassa-122	81	5	,	,	PUNCT
ijassa-122	81	6	w	w	NOUN
ijassa-122	81	7	)	)	PUNCT
ijassa-122	81	8	=	=	SYM
ijassa-122	81	9	2π−n/2(detγi)1/2	2π−n/2(detγi)1/2	NUM
ijassa-122	81	10	exp{−1	exp{−1	ADP
ijassa-122	81	11	2	2	NUM
ijassa-122	81	12	(	(	PUNCT
ijassa-122	81	13	z	z	NOUN
ijassa-122	81	14	−ψw	−ψw	NOUN
ijassa-122	81	15	)	)	PUNCT
ijassa-122	81	16	′γi(z	′γi(z	NOUN
ijassa-122	81	17	−ψw	−ψw	NOUN
ijassa-122	81	18	)	)	PUNCT
ijassa-122	81	19	}	}	PUNCT
ijassa-122	81	20	(	(	PUNCT
ijassa-122	81	21	3	3	X
ijassa-122	81	22	)	)	PUNCT
ijassa-122	81	23	the	the	DET
ijassa-122	81	24	joint	joint	ADJ
ijassa-122	81	25	maximum	maximum	ADJ
ijassa-122	81	26	likelihood	likelihood	NOUN
ijassa-122	81	27	estimate	estimate	NOUN
ijassa-122	81	28	w	w	NOUN
ijassa-122	81	29	∗	∗	NOUN
ijassa-122	81	30	of	of	ADP
ijassa-122	81	31	w	w	NOUN
ijassa-122	81	32	,	,	PUNCT
ijassa-122	81	33	given	give	VERB
ijassa-122	81	34	the	the	DET
ijassa-122	81	35	pattern	pattern	NOUN
ijassa-122	81	36	of	of	ADP
ijassa-122	81	37	the	the	DET
ijassa-122	81	38	outlying	outlying	ADJ
ijassa-122	81	39	observations	observation	NOUN
ijassa-122	81	40	and	and	CCONJ
ijassa-122	81	41	γi	γi	NOUN
ijassa-122	81	42	,	,	PUNCT
ijassa-122	81	43	is	be	AUX
ijassa-122	81	44	w	w	NOUN
ijassa-122	81	45	∗	∗	NOUN
ijassa-122	81	46	=	=	SYM
ijassa-122	81	47	(	(	PUNCT
ijassa-122	81	48	ψ′γiψ)−1ψ′γiz	ψ′γiψ)−1ψ′γiz	X
ijassa-122	81	49	and	and	CCONJ
ijassa-122	81	50	y	y	PROPN
ijassa-122	81	51	∗	∗	NOUN
ijassa-122	81	52	=	=	PUNCT
ijassa-122	81	53	z	z	NOUN
ijassa-122	81	54	−ψw	−ψw	NOUN
ijassa-122	81	55	∗	∗	NOUN
ijassa-122	81	56	(	(	PUNCT
ijassa-122	81	57	4	4	X
ijassa-122	81	58	)	)	PUNCT
ijassa-122	81	59	advances	advance	NOUN
ijassa-122	81	60	in	in	ADP
ijassa-122	81	61	systems	system	NOUN
ijassa-122	81	62	science	science	NOUN
ijassa-122	81	63	and	and	CCONJ
ijassa-122	81	64	applications	application	NOUN
ijassa-122	81	65	(	(	PUNCT
ijassa-122	81	66	2012	2012	NUM
ijassa-122	81	67	)	)	PUNCT
ijassa-122	82	1	vol.12	vol.12	NOUN
ijassa-122	82	2	no.4	no.4	PROPN
ijassa-122	82	3	403	403	NUM
ijassa-122	82	4	in	in	ADP
ijassa-122	82	5	practice	practice	NOUN
ijassa-122	82	6	,	,	PUNCT
ijassa-122	82	7	however	however	ADV
ijassa-122	82	8	,	,	PUNCT
ijassa-122	82	9	γi	γi	X
ijassa-122	82	10	is	be	AUX
ijassa-122	82	11	seldom	seldom	ADV
ijassa-122	82	12	known	know	VERB
ijassa-122	82	13	,	,	PUNCT
ijassa-122	82	14	so	so	ADV
ijassa-122	82	15	we	we	PRON
ijassa-122	82	16	have	have	VERB
ijassa-122	82	17	to	to	PART
ijassa-122	82	18	estimate	estimate	VERB
ijassa-122	82	19	it	it	PRON
ijassa-122	82	20	from	from	ADP
ijassa-122	82	21	the	the	DET
ijassa-122	82	22	data	datum	NOUN
ijassa-122	82	23	.	.	PUNCT
ijassa-122	83	1	similar	similar	ADJ
ijassa-122	83	2	to	to	ADP
ijassa-122	83	3	baragona	baragona	VERB
ijassa-122	83	4	et	et	NOUN
ijassa-122	83	5	al.[1	al.[1	PROPN
ijassa-122	83	6	]	]	PUNCT
ijassa-122	83	7	,	,	PUNCT
ijassa-122	83	8	using	use	VERB
ijassa-122	83	9	the	the	DET
ijassa-122	83	10	interpolator	interpolator	NOUN
ijassa-122	83	11	estimate	estimate	NOUN
ijassa-122	83	12	of	of	ADP
ijassa-122	83	13	yt	yt	PROPN
ijassa-122	83	14	.	.	PUNCT
ijassa-122	84	1	we	we	PRON
ijassa-122	84	2	may	may	AUX
ijassa-122	84	3	obtain	obtain	VERB
ijassa-122	84	4	inverse	inverse	ADJ
ijassa-122	84	5	autocorrelations	autocorrelation	NOUN
ijassa-122	84	6	interpolation	interpolation	NOUN
ijassa-122	84	7	estimates	estimate	VERB
ijassa-122	84	8	ρ̂ij	ρ̂ij	NOUN
ijassa-122	84	9	,	,	PUNCT
ijassa-122	84	10	j	j	PROPN
ijassa-122	84	11	=	=	SYM
ijassa-122	84	12	1	1	NUM
ijassa-122	84	13	,	,	PUNCT
ijassa-122	84	14	·	·	PUNCT
ijassa-122	84	15	·	·	PUNCT
ijassa-122	84	16	·	·	PUNCT
ijassa-122	84	17	,	,	PUNCT
ijassa-122	84	18	m	m	VERB
ijassa-122	84	19	and	and	CCONJ
ijassa-122	84	20	the	the	DET
ijassa-122	84	21	estimate	estimate	NOUN
ijassa-122	84	22	γ̂i0	γ̂i0	PROPN
ijassa-122	84	23	of	of	ADP
ijassa-122	84	24	inverse	inverse	ADJ
ijassa-122	84	25	variance	variance	NOUN
ijassa-122	84	26	and	and	CCONJ
ijassa-122	84	27	the	the	DET
ijassa-122	84	28	estimates	estimate	NOUN
ijassa-122	84	29	of	of	ADP
ijassa-122	84	30	inverse	inverse	NOUN
ijassa-122	84	31	autocovariance	autocovariance	NOUN
ijassa-122	84	32	γ̂ij	γ̂ij	NOUN
ijassa-122	84	33	=	=	SYM
ijassa-122	84	34	γ̂i0	γ̂i0	PUNCT
ijassa-122	84	35	·	·	PUNCT
ijassa-122	84	36	ρ̂ij	ρ̂ij	NOUN
ijassa-122	84	37	,	,	PUNCT
ijassa-122	84	38	j	j	PROPN
ijassa-122	84	39	=	=	SYM
ijassa-122	84	40	1	1	NUM
ijassa-122	84	41	,	,	PUNCT
ijassa-122	84	42	·	·	PUNCT
ijassa-122	84	43	·	·	PUNCT
ijassa-122	84	44	·	·	PUNCT
ijassa-122	84	45	,	,	PUNCT
ijassa-122	84	46	m.	m.	NOUN
ijassa-122	84	47	once	once	SCONJ
ijassa-122	84	48	the	the	DET
ijassa-122	84	49	inverse	inverse	NOUN
ijassa-122	84	50	autocovariances	autocovariance	NOUN
ijassa-122	84	51	have	have	AUX
ijassa-122	84	52	been	be	AUX
ijassa-122	84	53	estimated	estimate	VERB
ijassa-122	84	54	,	,	PUNCT
ijassa-122	84	55	the	the	DET
ijassa-122	84	56	ψj	ψj	ADV
ijassa-122	84	57	,	,	PUNCT
ijassa-122	84	58	j	j	PROPN
ijassa-122	84	59	=	=	SYM
ijassa-122	84	60	1	1	NUM
ijassa-122	84	61	,	,	PUNCT
ijassa-122	84	62	·	·	PUNCT
ijassa-122	84	63	·	·	PUNCT
ijassa-122	84	64	·	·	PUNCT
ijassa-122	84	65	,	,	PUNCT
ijassa-122	84	66	q	q	PROPN
ijassa-122	84	67	may	may	AUX
ijassa-122	84	68	be	be	AUX
ijassa-122	84	69	computed	compute	VERB
ijassa-122	84	70	using	use	VERB
ijassa-122	84	71	equation	equation	NOUN
ijassa-122	84	72	(	(	PUNCT
ijassa-122	84	73	2	2	NUM
ijassa-122	84	74	)	)	PUNCT
ijassa-122	84	75	.	.	PUNCT
ijassa-122	85	1	nevertheless	nevertheless	ADV
ijassa-122	85	2	,	,	PUNCT
ijassa-122	85	3	these	these	DET
ijassa-122	85	4	estimates	estimate	NOUN
ijassa-122	85	5	are	be	AUX
ijassa-122	85	6	biased	bias	VERB
ijassa-122	85	7	because	because	SCONJ
ijassa-122	85	8	of	of	ADP
ijassa-122	85	9	the	the	DET
ijassa-122	85	10	presence	presence	NOUN
ijassa-122	85	11	of	of	ADP
ijassa-122	85	12	the	the	DET
ijassa-122	85	13	outliers	outlier	NOUN
ijassa-122	85	14	.	.	PUNCT
ijassa-122	86	1	so	so	ADV
ijassa-122	86	2	we	we	PRON
ijassa-122	86	3	have	have	VERB
ijassa-122	86	4	to	to	PART
ijassa-122	86	5	resort	resort	VERB
ijassa-122	86	6	to	to	ADP
ijassa-122	86	7	an	an	DET
ijassa-122	86	8	iterative	iterative	NOUN
ijassa-122	86	9	scheme	scheme	NOUN
ijassa-122	86	10	for	for	ADP
ijassa-122	86	11	(	(	PUNCT
ijassa-122	86	12	4	4	NUM
ijassa-122	86	13	)	)	PUNCT
ijassa-122	86	14	,	,	PUNCT
ijassa-122	86	15	which	which	PRON
ijassa-122	86	16	is	be	AUX
ijassa-122	86	17	repeated	repeat	VERB
ijassa-122	86	18	until	until	ADP
ijassa-122	86	19	convergences	convergence	NOUN
ijassa-122	86	20	.	.	PUNCT
ijassa-122	87	1	if	if	SCONJ
ijassa-122	87	2	convergence	convergence	NOUN
ijassa-122	87	3	is	be	AUX
ijassa-122	87	4	attained	attain	VERB
ijassa-122	87	5	,	,	PUNCT
ijassa-122	87	6	and	and	CCONJ
ijassa-122	87	7	since	since	SCONJ
ijassa-122	87	8	the	the	DET
ijassa-122	87	9	likelihood	likelihood	NOUN
ijassa-122	87	10	is	be	AUX
ijassa-122	87	11	maximized	maximize	VERB
ijassa-122	87	12	with	with	ADP
ijassa-122	87	13	respect	respect	NOUN
ijassa-122	87	14	to	to	ADP
ijassa-122	87	15	the	the	DET
ijassa-122	87	16	inverse	inverse	NOUN
ijassa-122	87	17	autocorrelations	autocorrelation	NOUN
ijassa-122	87	18	,	,	PUNCT
ijassa-122	87	19	the	the	DET
ijassa-122	87	20	argument	argument	NOUN
ijassa-122	87	21	of	of	ADP
ijassa-122	87	22	the	the	DET
ijassa-122	87	23	exponential	exponential	NOUN
ijassa-122	87	24	in	in	ADP
ijassa-122	87	25	likelihood	likelihood	NOUN
ijassa-122	87	26	’s	’s	PART
ijassa-122	87	27	formula	formula	NOUN
ijassa-122	87	28	(	(	PUNCT
ijassa-122	87	29	3	3	X
ijassa-122	87	30	)	)	PUNCT
ijassa-122	87	31	is	be	AUX
ijassa-122	87	32	simply	simply	ADV
ijassa-122	87	33	−n/2	−n/2	ADJ
ijassa-122	87	34	,	,	PUNCT
ijassa-122	87	35	which	which	PRON
ijassa-122	87	36	is	be	AUX
ijassa-122	87	37	shown	show	VERB
ijassa-122	87	38	by	by	ADP
ijassa-122	87	39	the	the	DET
ijassa-122	87	40	following	follow	VERB
ijassa-122	87	41	proposition(the	proposition(the	ADJ
ijassa-122	87	42	proof	proof	NOUN
ijassa-122	87	43	is	be	AUX
ijassa-122	87	44	omited	omit	VERB
ijassa-122	87	45	)	)	PUNCT
ijassa-122	87	46	.	.	PUNCT
ijassa-122	88	1	proposition	proposition	NOUN
ijassa-122	88	2	1	1	NUM
ijassa-122	88	3	using	use	VERB
ijassa-122	88	4	the	the	DET
ijassa-122	88	5	former	former	ADJ
ijassa-122	88	6	representation	representation	NOUN
ijassa-122	88	7	and	and	CCONJ
ijassa-122	88	8	estimations	estimation	NOUN
ijassa-122	88	9	,	,	PUNCT
ijassa-122	88	10	then	then	ADV
ijassa-122	88	11	we	we	PRON
ijassa-122	88	12	have	have	VERB
ijassa-122	88	13	−1	−1	ADV
ijassa-122	88	14	2	2	NUM
ijassa-122	88	15	(	(	PUNCT
ijassa-122	88	16	z	z	NOUN
ijassa-122	88	17	−ψw	−ψw	NOUN
ijassa-122	88	18	)	)	PUNCT
ijassa-122	88	19	′γ(z	′γ(z	NOUN
ijassa-122	88	20	−ψw	−ψw	NOUN
ijassa-122	88	21	)	)	PUNCT
ijassa-122	88	22	=	=	SYM
ijassa-122	89	1	−1	−1	NOUN
ijassa-122	89	2	2	2	NUM
ijassa-122	89	3	n	n	CCONJ
ijassa-122	89	4	therefore	therefore	ADV
ijassa-122	89	5	,	,	PUNCT
ijassa-122	89	6	we	we	PRON
ijassa-122	89	7	have	have	VERB
ijassa-122	89	8	logl	logl	NOUN
ijassa-122	89	9	=	=	SYM
ijassa-122	89	10	−n	−n	ADJ
ijassa-122	89	11	2	2	NUM
ijassa-122	89	12	log(2π	log(2π	NOUN
ijassa-122	89	13	)	)	PUNCT
ijassa-122	90	1	+	+	CCONJ
ijassa-122	90	2	1	1	NUM
ijassa-122	90	3	2	2	NUM
ijassa-122	90	4	log(detγi)−	log(detγi)−	NOUN
ijassa-122	90	5	1	1	NUM
ijassa-122	90	6	2	2	NUM
ijassa-122	90	7	n	n	CCONJ
ijassa-122	90	8	thus	thus	ADV
ijassa-122	90	9	,	,	PUNCT
ijassa-122	90	10	the	the	DET
ijassa-122	90	11	likelihood	likelihood	NOUN
ijassa-122	90	12	function	function	NOUN
ijassa-122	90	13	basically	basically	ADV
ijassa-122	90	14	depends	depend	VERB
ijassa-122	90	15	on	on	ADP
ijassa-122	90	16	detγi	detγi	NOUN
ijassa-122	90	17	.	.	PUNCT
ijassa-122	91	1	because	because	SCONJ
ijassa-122	91	2	that	that	SCONJ
ijassa-122	91	3	the	the	DET
ijassa-122	91	4	fitness	fitness	NOUN
ijassa-122	91	5	function(ff	function(ff	PROPN
ijassa-122	91	6	)	)	PUNCT
ijassa-122	91	7	has	have	VERB
ijassa-122	91	8	to	to	PART
ijassa-122	91	9	be	be	AUX
ijassa-122	91	10	positive	positive	ADJ
ijassa-122	91	11	,	,	PUNCT
ijassa-122	91	12	we	we	PRON
ijassa-122	91	13	let	let	VERB
ijassa-122	91	14	f	f	PROPN
ijassa-122	91	15	(	(	PUNCT
ijassa-122	91	16	ξ	ξ	PROPN
ijassa-122	91	17	)	)	PUNCT
ijassa-122	91	18	=	=	SYM
ijassa-122	91	19	a	a	PRON
ijassa-122	91	20	·	·	PUNCT
ijassa-122	91	21	blog(detγi)−ck	blog(detγi)−ck	NOUN
ijassa-122	91	22	,	,	PUNCT
ijassa-122	91	23	where	where	SCONJ
ijassa-122	91	24	b	b	NOUN
ijassa-122	91	25	is	be	AUX
ijassa-122	91	26	a	a	DET
ijassa-122	91	27	real	real	ADV
ijassa-122	91	28	constant	constant	ADJ
ijassa-122	91	29	such	such	ADJ
ijassa-122	91	30	that	that	DET
ijassa-122	91	31	b	b	PROPN
ijassa-122	91	32	>	>	X
ijassa-122	91	33	1	1	NUM
ijassa-122	91	34	,	,	PUNCT
ijassa-122	91	35	and	and	CCONJ
ijassa-122	91	36	the	the	DET
ijassa-122	91	37	constant	constant	ADJ
ijassa-122	91	38	terms	term	NOUN
ijassa-122	91	39	in	in	ADP
ijassa-122	91	40	the	the	DET
ijassa-122	91	41	exponent	exponent	NOUN
ijassa-122	91	42	were	be	AUX
ijassa-122	91	43	dropped	drop	VERB
ijassa-122	91	44	.	.	PUNCT
ijassa-122	92	1	we	we	PRON
ijassa-122	92	2	use	use	VERB
ijassa-122	92	3	a	a	DET
ijassa-122	92	4	=	=	SYM
ijassa-122	92	5	10	10	NUM
ijassa-122	92	6	,	,	PUNCT
ijassa-122	92	7	b	b	NOUN
ijassa-122	92	8	=	=	SYM
ijassa-122	92	9	1.0001	1.0001	PROPN
ijassa-122	92	10	and	and	CCONJ
ijassa-122	92	11	c	c	NOUN
ijassa-122	92	12	=	=	SYM
ijassa-122	92	13	14	14	NUM
ijassa-122	92	14	in	in	ADP
ijassa-122	92	15	this	this	DET
ijassa-122	92	16	paper	paper	NOUN
ijassa-122	92	17	.	.	PUNCT
ijassa-122	93	1	we	we	PRON
ijassa-122	93	2	adopted	adopt	VERB
ijassa-122	93	3	(	(	PUNCT
ijassa-122	93	4	1	1	NUM
ijassa-122	93	5	)	)	PUNCT
ijassa-122	93	6	population	population	NOUN
ijassa-122	93	7	size	size	NOUN
ijassa-122	93	8	s	s	PART
ijassa-122	93	9	=	=	SYM
ijassa-122	93	10	n	n	PROPN
ijassa-122	93	11	+	+	NOUN
ijassa-122	93	12	1	1	NUM
ijassa-122	93	13	;	;	PUNCT
ijassa-122	93	14	(	(	PUNCT
ijassa-122	93	15	2	2	X
ijassa-122	93	16	)	)	PUNCT
ijassa-122	93	17	probability	probability	NOUN
ijassa-122	93	18	of	of	ADP
ijassa-122	93	19	crossover	crossover	NOUN
ijassa-122	93	20	pc	pc	NOUN
ijassa-122	93	21	=	=	NUM
ijassa-122	93	22	0.8	0.8	NUM
ijassa-122	93	23	;	;	PUNCT
ijassa-122	93	24	(	(	PUNCT
ijassa-122	93	25	3	3	X
ijassa-122	93	26	)	)	PUNCT
ijassa-122	93	27	probability	probability	NOUN
ijassa-122	93	28	of	of	ADP
ijassa-122	93	29	mutation	mutation	NOUN
ijassa-122	93	30	pm	pm	NOUN
ijassa-122	93	31	=	=	NOUN
ijassa-122	93	32	0.05	0.05	NUM
ijassa-122	93	33	;	;	PUNCT
ijassa-122	93	34	(	(	PUNCT
ijassa-122	93	35	4	4	X
ijassa-122	93	36	)	)	PUNCT
ijassa-122	93	37	the	the	DET
ijassa-122	93	38	maximum	maximum	ADJ
ijassa-122	93	39	number	number	NOUN
ijassa-122	93	40	of	of	ADP
ijassa-122	93	41	outliers	outlier	NOUN
ijassa-122	93	42	within	within	ADP
ijassa-122	93	43	a	a	DET
ijassa-122	93	44	chromosome	chromosome	NOUN
ijassa-122	93	45	g	g	NOUN
ijassa-122	93	46	=	=	NOUN
ijassa-122	93	47	10	10	NUM
ijassa-122	93	48	;	;	PUNCT
ijassa-122	93	49	(	(	PUNCT
ijassa-122	93	50	5	5	X
ijassa-122	93	51	)	)	PUNCT
ijassa-122	93	52	the	the	DET
ijassa-122	93	53	number	number	NOUN
ijassa-122	93	54	of	of	ADP
ijassa-122	93	55	iterations	iteration	NOUN
ijassa-122	93	56	of	of	ADP
ijassa-122	93	57	the	the	DET
ijassa-122	93	58	series	series	NOUN
ijassa-122	93	59	’	'	PUNCT
ijassa-122	93	60	adjustment	adjustment	NOUN
ijassa-122	93	61	/	/	SYM
ijassa-122	93	62	parameters	parameter	NOUN
ijassa-122	93	63	’	'	PUNCT
ijassa-122	93	64	estimation	estimation	NOUN
ijassa-122	93	65	procedure	procedure	NOUN
ijassa-122	93	66	was	be	AUX
ijassa-122	93	67	3;(6	3;(6	NUM
ijassa-122	93	68	)	)	PUNCT
ijassa-122	93	69	we	we	PRON
ijassa-122	93	70	perform	perform	VERB
ijassa-122	93	71	as	as	ADV
ijassa-122	93	72	many	many	ADJ
ijassa-122	93	73	iterations	iteration	NOUN
ijassa-122	93	74	of	of	ADP
ijassa-122	93	75	the	the	DET
ijassa-122	93	76	genetic	genetic	ADJ
ijassa-122	93	77	algorithm	algorithm	NOUN
ijassa-122	93	78	as	as	ADP
ijassa-122	93	79	possible	possible	ADJ
ijassa-122	93	80	within	within	ADP
ijassa-122	93	81	a	a	DET
ijassa-122	93	82	reasonable	reasonable	ADJ
ijassa-122	93	83	time	time	NOUN
ijassa-122	93	84	period	period	NOUN
ijassa-122	93	85	.	.	PUNCT
ijassa-122	94	1	5	5	NUM
ijassa-122	94	2	simulation	simulation	NOUN
ijassa-122	94	3	studies	study	NOUN
ijassa-122	94	4	example	example	VERB
ijassa-122	94	5	a	a	PRON
ijassa-122	94	6	in	in	ADP
ijassa-122	94	7	the	the	DET
ijassa-122	94	8	following	follow	VERB
ijassa-122	94	9	example	example	NOUN
ijassa-122	94	10	,	,	PUNCT
ijassa-122	94	11	we	we	PRON
ijassa-122	94	12	consider	consider	VERB
ijassa-122	94	13	the	the	DET
ijassa-122	94	14	model	model	NOUN
ijassa-122	94	15	(	(	PUNCT
ijassa-122	94	16	1−	1−	NUM
ijassa-122	94	17	0.8b	0.8b	NOUN
ijassa-122	94	18	+	+	CCONJ
ijassa-122	94	19	0.3b2)xt	0.3b2)xt	X
ijassa-122	95	1	=	=	PUNCT
ijassa-122	95	2	εt	εt	PROPN
ijassa-122	95	3	zt	zt	PROPN
ijassa-122	95	4	=	=	PRON
ijassa-122	95	5	(	(	PUNCT
ijassa-122	95	6	1−	1−	NUM
ijassa-122	95	7	0.5b)xt	0.5b)xt	PROPN
ijassa-122	95	8	−	−	ADP
ijassa-122	96	1	4δt,30	4δt,30	NUM
ijassa-122	97	1	+	+	CCONJ
ijassa-122	98	1	5δt,31	5δt,31	NUM
ijassa-122	98	2	−	−	NOUN
ijassa-122	98	3	5δt,80	5δt,80	NUM
ijassa-122	98	4	−	−	NOUN
ijassa-122	98	5	4δt,90	4δt,90	NUM
ijassa-122	98	6	+	+	SYM
ijassa-122	98	7	6×	6×	NOUN
ijassa-122	98	8	1−	1−	NUM
ijassa-122	98	9	0.36b	0.36b	NUM
ijassa-122	99	1	+	+	CCONJ
ijassa-122	99	2	0.85b2	0.85b2	PROPN
ijassa-122	99	3	1−	1−	NUM
ijassa-122	99	4	0.6b	0.6b	NOUN
ijassa-122	99	5	δt,40	δt,40	NOUN
ijassa-122	100	1	+	+	CCONJ
ijassa-122	100	2	1−	1−	NUM
ijassa-122	100	3	0.36b	0.36b	NUM
ijassa-122	100	4	+	+	CCONJ
ijassa-122	100	5	0.85b2	0.85b2	PROPN
ijassa-122	100	6	1−	1−	NUM
ijassa-122	100	7	0.6b	0.6b	NOUN
ijassa-122	100	8	et	et	NOUN
ijassa-122	100	9	where	where	SCONJ
ijassa-122	100	10	{	{	PUNCT
ijassa-122	100	11	εt	εt	X
ijassa-122	100	12	}	}	PUNCT
ijassa-122	100	13	and	and	CCONJ
ijassa-122	100	14	{	{	PUNCT
ijassa-122	100	15	et	et	NOUN
ijassa-122	100	16	}	}	PUNCT
ijassa-122	100	17	are	be	AUX
ijassa-122	100	18	all	all	PRON
ijassa-122	100	19	normal	normal	ADJ
ijassa-122	100	20	white	white	ADJ
ijassa-122	100	21	noise	noise	NOUN
ijassa-122	100	22	,	,	PUNCT
ijassa-122	100	23	their	their	PRON
ijassa-122	100	24	means	mean	NOUN
ijassa-122	100	25	are	be	AUX
ijassa-122	100	26	zero	zero	NUM
ijassa-122	100	27	and	and	CCONJ
ijassa-122	100	28	variance	variance	NOUN
ijassa-122	100	29	σ2	σ2	NOUN
ijassa-122	100	30	=	=	NOUN
ijassa-122	100	31	1	1	X
ijassa-122	100	32	.	.	X
ijassa-122	101	1	we	we	PRON
ijassa-122	101	2	create	create	VERB
ijassa-122	101	3	101	101	NUM
ijassa-122	101	4	observations	observation	NOUN
ijassa-122	101	5	x0	x0	PROPN
ijassa-122	101	6	,	,	PUNCT
ijassa-122	101	7	x1	x1	PROPN
ijassa-122	101	8	,	,	PUNCT
ijassa-122	101	9	·	·	PUNCT
ijassa-122	101	10	·	·	PUNCT
ijassa-122	101	11	·	·	PUNCT
ijassa-122	101	12	,	,	PUNCT
ijassa-122	101	13	x100	x100	PROPN
ijassa-122	101	14	of	of	ADP
ijassa-122	101	15	xt	xt	PROPN
ijassa-122	101	16	and	and	CCONJ
ijassa-122	101	17	100	100	NUM
ijassa-122	101	18	observations	observation	NOUN
ijassa-122	101	19	z1,	z1,	NOUN
ijassa-122	101	20	...	...	PUNCT
ijassa-122	101	21	,z100	,z100	PUNCT
ijassa-122	101	22	of	of	ADP
ijassa-122	101	23	zt	zt	PROPN
ijassa-122	101	24	by	by	ADP
ijassa-122	101	25	simulation	simulation	NOUN
ijassa-122	101	26	.	.	PUNCT
ijassa-122	102	1	obviously	obviously	ADV
ijassa-122	102	2	,	,	PUNCT
ijassa-122	102	3	it	it	PRON
ijassa-122	102	4	is	be	AUX
ijassa-122	102	5	ao	ao	PROPN
ijassa-122	102	6	at	at	ADP
ijassa-122	102	7	t	t	PROPN
ijassa-122	102	8	=	=	SYM
ijassa-122	102	9	30	30	NUM
ijassa-122	102	10	,	,	PUNCT
ijassa-122	102	11	31	31	NUM
ijassa-122	102	12	,	,	PUNCT
ijassa-122	102	13	80	80	NUM
ijassa-122	102	14	,	,	PUNCT
ijassa-122	102	15	90	90	NUM
ijassa-122	102	16	singly	singly	ADV
ijassa-122	102	17	and	and	CCONJ
ijassa-122	102	18	io	io	X
ijassa-122	102	19	at	at	ADP
ijassa-122	102	20	404	404	NUM
ijassa-122	102	21	ping	ping	NOUN
ijassa-122	102	22	chen	chen	PROPN
ijassa-122	102	23	:	:	PUNCT
ijassa-122	102	24	the	the	DET
ijassa-122	102	25	identification	identification	NOUN
ijassa-122	102	26	of	of	ADP
ijassa-122	102	27	outliers	outlier	NOUN
ijassa-122	102	28	in	in	ADP
ijassa-122	102	29	armax	armax	NOUN
ijassa-122	102	30	models	model	NOUN
ijassa-122	102	31	via	via	ADP
ijassa-122	102	32	genetic	genetic	ADJ
ijassa-122	102	33	algorithm	algorithm	NOUN
ijassa-122	102	34	t	t	NOUN
ijassa-122	102	35	=	=	SYM
ijassa-122	102	36	40	40	NUM
ijassa-122	102	37	,	,	PUNCT
ijassa-122	102	38	and	and	CCONJ
ijassa-122	102	39	outlier	outlier	NOUN
ijassa-122	102	40	magnitudes	magnitude	NOUN
ijassa-122	102	41	are	be	AUX
ijassa-122	102	42	w30	w30	NOUN
ijassa-122	102	43	=	=	SYM
ijassa-122	102	44	−4	−4	PROPN
ijassa-122	102	45	,	,	PUNCT
ijassa-122	102	46	w31	w31	NOUN
ijassa-122	102	47	=	=	SYM
ijassa-122	102	48	5	5	NUM
ijassa-122	102	49	,	,	PUNCT
ijassa-122	102	50	w80	w80	AUX
ijassa-122	102	51	=	=	SYM
ijassa-122	102	52	−5	−5	ADJ
ijassa-122	102	53	,	,	PUNCT
ijassa-122	102	54	w90	w90	NOUN
ijassa-122	102	55	=	=	SYM
ijassa-122	102	56	−4	−4	X
ijassa-122	102	57	and	and	CCONJ
ijassa-122	102	58	w40	w40	NUM
ijassa-122	102	59	=	=	SYM
ijassa-122	102	60	6	6	NUM
ijassa-122	102	61	,	,	PUNCT
ijassa-122	102	62	respectively	respectively	ADV
ijassa-122	102	63	.	.	PUNCT
ijassa-122	103	1	applying	apply	VERB
ijassa-122	103	2	our	our	PRON
ijassa-122	103	3	method	method	NOUN
ijassa-122	103	4	to	to	ADP
ijassa-122	103	5	the	the	DET
ijassa-122	103	6	above	above	ADJ
ijassa-122	103	7	data	datum	NOUN
ijassa-122	103	8	and	and	CCONJ
ijassa-122	103	9	prewhitening	prewhitene	VERB
ijassa-122	103	10	the	the	DET
ijassa-122	103	11	input	input	NOUN
ijassa-122	103	12	series	series	NOUN
ijassa-122	103	13	.	.	PUNCT
ijassa-122	104	1	making	make	VERB
ijassa-122	104	2	{	{	PUNCT
ijassa-122	104	3	xt	xt	VERB
ijassa-122	104	4	}	}	PUNCT
ijassa-122	104	5	follows	follow	VERB
ijassa-122	104	6	an	an	DET
ijassa-122	104	7	arma	arma	PROPN
ijassa-122	104	8	model	model	NOUN
ijassa-122	104	9	:	:	PUNCT
ijassa-122	104	10	(	(	PUNCT
ijassa-122	104	11	1−	1−	NUM
ijassa-122	104	12	0.88088b	0.88088b	NOUN
ijassa-122	104	13	+	+	CCONJ
ijassa-122	104	14	0.32738b2)xt	0.32738b2)xt	X
ijassa-122	104	15	=	=	SYM
ijassa-122	105	1	εt	εt	PROPN
ijassa-122	105	2	.	.	PUNCT
ijassa-122	106	1	then	then	ADV
ijassa-122	106	2	we	we	PRON
ijassa-122	106	3	take	take	VERB
ijassa-122	106	4	the	the	DET
ijassa-122	106	5	same	same	ADJ
ijassa-122	106	6	manipulation	manipulation	NOUN
ijassa-122	106	7	to	to	PART
ijassa-122	106	8	prewhiten	prewhiten	VERB
ijassa-122	106	9	{	{	PUNCT
ijassa-122	106	10	zt	zt	PROPN
ijassa-122	106	11	}	}	PUNCT
ijassa-122	106	12	.	.	PUNCT
ijassa-122	107	1	by	by	ADP
ijassa-122	107	2	analyzing	analyze	VERB
ijassa-122	107	3	filtered	filter	VERB
ijassa-122	107	4	cross	cross	NOUN
ijassa-122	107	5	correlation	correlation	NOUN
ijassa-122	107	6	coefficient	coefficient	NOUN
ijassa-122	107	7	of	of	ADP
ijassa-122	107	8	{	{	PUNCT
ijassa-122	107	9	zt	zt	PROPN
ijassa-122	107	10	}	}	PUNCT
ijassa-122	107	11	and	and	CCONJ
ijassa-122	107	12	{	{	PUNCT
ijassa-122	107	13	xt	xt	ADP
ijassa-122	107	14	}	}	PUNCT
ijassa-122	107	15	,	,	PUNCT
ijassa-122	107	16	we	we	PRON
ijassa-122	107	17	obtain	obtain	VERB
ijassa-122	107	18	the	the	DET
ijassa-122	107	19	transfer	transfer	NOUN
ijassa-122	107	20	function	function	NOUN
ijassa-122	107	21	1.03416−0.73967b	1.03416−0.73967b	NUM
ijassa-122	107	22	for	for	ADP
ijassa-122	107	23	{	{	PUNCT
ijassa-122	107	24	xt	xt	ADP
ijassa-122	107	25	}	}	PUNCT
ijassa-122	107	26	.	.	PUNCT
ijassa-122	108	1	delete	delete	VERB
ijassa-122	108	2	the	the	DET
ijassa-122	108	3	influence	influence	NOUN
ijassa-122	108	4	of	of	ADP
ijassa-122	108	5	input	input	NOUN
ijassa-122	108	6	process	process	NOUN
ijassa-122	108	7	{	{	PUNCT
ijassa-122	108	8	xt	xt	ADP
ijassa-122	108	9	}	}	PUNCT
ijassa-122	108	10	in	in	ADP
ijassa-122	108	11	response	response	NOUN
ijassa-122	108	12	process	process	NOUN
ijassa-122	108	13	{	{	PUNCT
ijassa-122	108	14	zt	zt	PROPN
ijassa-122	108	15	}	}	PUNCT
ijassa-122	108	16	,	,	PUNCT
ijassa-122	108	17	and	and	CCONJ
ijassa-122	108	18	let	let	VERB
ijassa-122	108	19	z∗t	z∗t	NUM
ijassa-122	108	20	=	=	SYM
ijassa-122	108	21	zt	zt	PROPN
ijassa-122	108	22	−	−	PROPN
ijassa-122	108	23	(	(	PUNCT
ijassa-122	108	24	1.03416	1.03416	NUM
ijassa-122	108	25	−	−	PROPN
ijassa-122	108	26	0.73967b)xt	0.73967b)xt	PROPN
ijassa-122	108	27	.	.	PUNCT
ijassa-122	109	1	we	we	PRON
ijassa-122	109	2	have	have	VERB
ijassa-122	109	3	that	that	SCONJ
ijassa-122	109	4	{	{	PUNCT
ijassa-122	109	5	z∗t	z∗t	NUM
ijassa-122	109	6	}	}	PUNCT
ijassa-122	109	7	is	be	AUX
ijassa-122	109	8	an	an	DET
ijassa-122	109	9	arma	arma	PROPN
ijassa-122	109	10	series	series	NOUN
ijassa-122	109	11	include	include	VERB
ijassa-122	109	12	outliers	outlier	NOUN
ijassa-122	109	13	.	.	PUNCT
ijassa-122	110	1	we	we	PRON
ijassa-122	110	2	detect	detect	VERB
ijassa-122	110	3	the	the	DET
ijassa-122	110	4	outliers	outlier	NOUN
ijassa-122	110	5	in	in	ADP
ijassa-122	110	6	{	{	PUNCT
ijassa-122	110	7	z∗t	z∗t	NUM
ijassa-122	110	8	}	}	PUNCT
ijassa-122	110	9	by	by	ADP
ijassa-122	110	10	applying	apply	VERB
ijassa-122	110	11	the	the	DET
ijassa-122	110	12	above	above	ADJ
ijassa-122	110	13	method	method	NOUN
ijassa-122	110	14	.	.	PUNCT
ijassa-122	111	1	let	let	VERB
ijassa-122	111	2	m	m	NOUN
ijassa-122	111	3	=	=	SYM
ijassa-122	111	4	2	2	NUM
ijassa-122	111	5	,	,	PUNCT
ijassa-122	111	6	q	q	NOUN
ijassa-122	111	7	=	=	NOUN
ijassa-122	111	8	9	9	NUM
ijassa-122	111	9	,	,	PUNCT
ijassa-122	111	10	g	g	NOUN
ijassa-122	111	11	=	=	SYM
ijassa-122	111	12	10	10	NUM
ijassa-122	111	13	,	,	PUNCT
ijassa-122	111	14	s	s	PART
ijassa-122	111	15	=	=	NOUN
ijassa-122	111	16	101	101	NUM
ijassa-122	111	17	.	.	PUNCT
ijassa-122	112	1	because	because	SCONJ
ijassa-122	112	2	the	the	DET
ijassa-122	112	3	30th	30th	ADJ
ijassa-122	112	4	point	point	NOUN
ijassa-122	112	5	is	be	AUX
ijassa-122	112	6	very	very	ADV
ijassa-122	112	7	close	close	ADJ
ijassa-122	112	8	to	to	ADP
ijassa-122	112	9	31th	31th	ADJ
ijassa-122	112	10	point	point	NOUN
ijassa-122	112	11	and	and	CCONJ
ijassa-122	112	12	they	they	PRON
ijassa-122	112	13	influence	influence	VERB
ijassa-122	112	14	each	each	DET
ijassa-122	112	15	other	other	ADJ
ijassa-122	112	16	,	,	PUNCT
ijassa-122	112	17	it	it	PRON
ijassa-122	112	18	is	be	AUX
ijassa-122	112	19	difficult	difficult	ADJ
ijassa-122	112	20	to	to	PART
ijassa-122	112	21	identify	identify	VERB
ijassa-122	112	22	the	the	DET
ijassa-122	112	23	outliers	outlier	NOUN
ijassa-122	112	24	.	.	PUNCT
ijassa-122	113	1	in	in	ADP
ijassa-122	113	2	this	this	DET
ijassa-122	113	3	case	case	NOUN
ijassa-122	113	4	,	,	PUNCT
ijassa-122	113	5	one	one	PRON
ijassa-122	113	6	needs	need	VERB
ijassa-122	113	7	larger	large	ADJ
ijassa-122	113	8	number	number	NOUN
ijassa-122	113	9	of	of	ADP
ijassa-122	113	10	iterations	iteration	NOUN
ijassa-122	113	11	.	.	PUNCT
ijassa-122	114	1	we	we	PRON
ijassa-122	114	2	take	take	VERB
ijassa-122	114	3	1000	1000	NUM
ijassa-122	114	4	iterations	iteration	NOUN
ijassa-122	114	5	by	by	ADP
ijassa-122	114	6	standard	standard	ADJ
ijassa-122	114	7	genetic	genetic	ADJ
ijassa-122	114	8	algorithm	algorithm	NOUN
ijassa-122	114	9	.	.	PUNCT
ijassa-122	115	1	and	and	CCONJ
ijassa-122	115	2	obtain	obtain	VERB
ijassa-122	115	3	the	the	DET
ijassa-122	115	4	best	good	ADJ
ijassa-122	115	5	individual	individual	NOUN
ijassa-122	115	6	at	at	ADP
ijassa-122	115	7	612th	612th	ADJ
ijassa-122	115	8	iterations	iteration	NOUN
ijassa-122	115	9	:	:	PUNCT
ijassa-122	115	10	t	t	PROPN
ijassa-122	115	11	=	=	SYM
ijassa-122	115	12	30(ao	30(ao	NUM
ijassa-122	115	13	)	)	PUNCT
ijassa-122	115	14	,	,	PUNCT
ijassa-122	115	15	w∗	w∗	NOUN
ijassa-122	115	16	30	30	NUM
ijassa-122	115	17	=	=	SYM
ijassa-122	115	18	−4.7172	−4.7172	NOUN
ijassa-122	115	19	;	;	PUNCT
ijassa-122	115	20	t	t	PROPN
ijassa-122	115	21	=	=	PUNCT
ijassa-122	115	22	31(ao	31(ao	NUM
ijassa-122	115	23	)	)	PUNCT
ijassa-122	115	24	,	,	PUNCT
ijassa-122	115	25	w∗	w∗	NOUN
ijassa-122	115	26	31	31	NUM
ijassa-122	115	27	=	=	SYM
ijassa-122	115	28	4.1010	4.1010	NUM
ijassa-122	115	29	;	;	PUNCT
ijassa-122	115	30	t	t	NOUN
ijassa-122	115	31	=	=	SYM
ijassa-122	115	32	40(io	40(io	NUM
ijassa-122	115	33	)	)	PUNCT
ijassa-122	115	34	,	,	PUNCT
ijassa-122	115	35	w∗	w∗	NOUN
ijassa-122	115	36	40	40	NUM
ijassa-122	115	37	=	=	SYM
ijassa-122	115	38	4.0487	4.0487	NUM
ijassa-122	115	39	;	;	PUNCT
ijassa-122	115	40	t	t	NOUN
ijassa-122	115	41	=	=	PUNCT
ijassa-122	115	42	80(ao	80(ao	NUM
ijassa-122	115	43	)	)	PUNCT
ijassa-122	115	44	,	,	PUNCT
ijassa-122	115	45	w∗	w∗	NOUN
ijassa-122	115	46	80	80	NUM
ijassa-122	115	47	=	=	SYM
ijassa-122	115	48	−5.6440	−5.6440	PROPN
ijassa-122	115	49	;	;	PUNCT
ijassa-122	115	50	t	t	PROPN
ijassa-122	115	51	=	=	SYM
ijassa-122	115	52	90(ao	90(ao	NUM
ijassa-122	115	53	)	)	PUNCT
ijassa-122	115	54	,	,	PUNCT
ijassa-122	115	55	w∗	w∗	NOUN
ijassa-122	115	56	90	90	NUM
ijassa-122	115	57	=	=	SYM
ijassa-122	115	58	−4.6228	−4.6228	PROPN
ijassa-122	115	59	the	the	DET
ijassa-122	115	60	outcome	outcome	NOUN
ijassa-122	115	61	is	be	AUX
ijassa-122	115	62	consistent	consistent	ADJ
ijassa-122	115	63	with	with	ADP
ijassa-122	115	64	our	our	PRON
ijassa-122	115	65	prearrangement	prearrangement	NOUN
ijassa-122	115	66	.	.	PUNCT
ijassa-122	116	1	the	the	DET
ijassa-122	116	2	outliers	outlier	NOUN
ijassa-122	116	3	in	in	ADP
ijassa-122	116	4	{	{	PUNCT
ijassa-122	116	5	zt	zt	PROPN
ijassa-122	116	6	}	}	PUNCT
ijassa-122	116	7	process	process	NOUN
ijassa-122	116	8	are	be	AUX
ijassa-122	116	9	detected	detect	VERB
ijassa-122	116	10	successfully	successfully	ADV
ijassa-122	116	11	,	,	PUNCT
ijassa-122	116	12	and	and	CCONJ
ijassa-122	116	13	there	there	PRON
ijassa-122	116	14	is	be	VERB
ijassa-122	116	15	no	no	DET
ijassa-122	116	16	misjudgement	misjudgement	NOUN
ijassa-122	116	17	.	.	PUNCT
ijassa-122	117	1	6	6	NUM
ijassa-122	117	2	conclusions	conclusion	NOUN
ijassa-122	117	3	there	there	PRON
ijassa-122	117	4	are	be	VERB
ijassa-122	117	5	ao	ao	NOUN
ijassa-122	117	6	and	and	CCONJ
ijassa-122	117	7	io	io	X
ijassa-122	117	8	in	in	ADP
ijassa-122	117	9	our	our	PRON
ijassa-122	117	10	model	model	NOUN
ijassa-122	117	11	,	,	PUNCT
ijassa-122	117	12	and	and	CCONJ
ijassa-122	117	13	also	also	ADV
ijassa-122	117	14	the	the	DET
ijassa-122	117	15	outliers(ao	outliers(ao	NOUN
ijassa-122	117	16	)	)	PUNCT
ijassa-122	117	17	present	present	ADJ
ijassa-122	117	18	consecutively	consecutively	ADV
ijassa-122	117	19	.	.	PUNCT
ijassa-122	118	1	it	it	PRON
ijassa-122	118	2	is	be	AUX
ijassa-122	118	3	quite	quite	ADV
ijassa-122	118	4	difficult	difficult	ADJ
ijassa-122	118	5	to	to	PART
ijassa-122	118	6	detect	detect	VERB
ijassa-122	118	7	outliers	outlier	NOUN
ijassa-122	118	8	,	,	PUNCT
ijassa-122	118	9	however	however	ADV
ijassa-122	118	10	,	,	PUNCT
ijassa-122	118	11	all	all	PRON
ijassa-122	118	12	of	of	ADP
ijassa-122	118	13	the	the	DET
ijassa-122	118	14	outliers	outlier	NOUN
ijassa-122	118	15	in	in	ADP
ijassa-122	118	16	the	the	DET
ijassa-122	118	17	above	above	ADJ
ijassa-122	118	18	model	model	NOUN
ijassa-122	118	19	have	have	AUX
ijassa-122	118	20	been	be	AUX
ijassa-122	118	21	detected	detect	VERB
ijassa-122	118	22	successfully	successfully	ADV
ijassa-122	118	23	,	,	PUNCT
ijassa-122	118	24	which	which	PRON
ijassa-122	118	25	shows	show	VERB
ijassa-122	118	26	our	our	PRON
ijassa-122	118	27	method	method	NOUN
ijassa-122	118	28	is	be	AUX
ijassa-122	118	29	also	also	ADV
ijassa-122	118	30	efficient	efficient	ADJ
ijassa-122	118	31	for	for	ADP
ijassa-122	118	32	the	the	DET
ijassa-122	118	33	ao	ao	PROPN
ijassa-122	118	34	and	and	CCONJ
ijassa-122	118	35	io	io	X
ijassa-122	118	36	problem	problem	NOUN
ijassa-122	118	37	in	in	ADP
ijassa-122	118	38	armax	armax	NOUN
ijassa-122	118	39	model	model	NOUN
ijassa-122	118	40	.	.	PUNCT
ijassa-122	119	1	some	some	DET
ijassa-122	119	2	other	other	ADJ
ijassa-122	119	3	case	case	NOUN
ijassa-122	119	4	studies	study	NOUN
ijassa-122	119	5	also	also	ADV
ijassa-122	119	6	show	show	VERB
ijassa-122	119	7	that	that	SCONJ
ijassa-122	119	8	our	our	PRON
ijassa-122	119	9	method	method	NOUN
ijassa-122	119	10	is	be	AUX
ijassa-122	119	11	effective	effective	ADJ
ijassa-122	119	12	in	in	ADP
ijassa-122	119	13	detecting	detect	VERB
ijassa-122	119	14	outliers	outlier	NOUN
ijassa-122	119	15	’	’	PART
ijassa-122	119	16	location	location	NOUN
ijassa-122	119	17	and	and	CCONJ
ijassa-122	119	18	type	type	NOUN
ijassa-122	119	19	and	and	CCONJ
ijassa-122	119	20	in	in	ADP
ijassa-122	119	21	estimating	estimate	VERB
ijassa-122	119	22	their	their	PRON
ijassa-122	119	23	size	size	NOUN
ijassa-122	119	24	for	for	ADP
ijassa-122	119	25	armax	armax	NOUN
ijassa-122	119	26	model	model	NOUN
ijassa-122	119	27	.	.	PUNCT
ijassa-122	120	1	acknowledgements	acknowledgement	NOUN
ijassa-122	120	2	the	the	DET
ijassa-122	120	3	research	research	NOUN
ijassa-122	120	4	is	be	AUX
ijassa-122	120	5	supported	support	VERB
ijassa-122	120	6	by	by	ADP
ijassa-122	120	7	national	national	ADJ
ijassa-122	120	8	natural	natural	PROPN
ijassa-122	120	9	science	science	PROPN
ijassa-122	120	10	foundation	foundation	PROPN
ijassa-122	120	11	of	of	ADP
ijassa-122	120	12	china	china	PROPN
ijassa-122	120	13	(	(	PUNCT
ijassa-122	120	14	no.11171065	no.11171065	PROPN
ijassa-122	120	15	)	)	PUNCT
ijassa-122	120	16	references	reference	NOUN
ijassa-122	121	1	[	[	X
ijassa-122	121	2	1	1	NUM
ijassa-122	121	3	]	]	X
ijassa-122	121	4	baragona	baragona	PROPN
ijassa-122	121	5	r	r	PROPN
ijassa-122	121	6	,	,	PUNCT
ijassa-122	121	7	battaglia	battaglia	PROPN
ijassa-122	121	8	f	f	PROPN
ijassa-122	121	9	,	,	PUNCT
ijassa-122	121	10	calzini	calzini	PROPN
ijassa-122	121	11	c.	c.	NOUN
ijassa-122	121	12	(	(	PUNCT
ijassa-122	121	13	2001	2001	NUM
ijassa-122	121	14	)	)	PUNCT
ijassa-122	121	15	,	,	PUNCT
ijassa-122	121	16	“	"	PUNCT
ijassa-122	121	17	genetic	genetic	ADJ
ijassa-122	121	18	algorithms	algorithm	NOUN
ijassa-122	121	19	for	for	ADP
ijassa-122	121	20	the	the	DET
ijassa-122	121	21	identification	identification	NOUN
ijassa-122	121	22	of	of	ADP
ijassa-122	121	23	additive	additive	NOUN
ijassa-122	121	24	and	and	CCONJ
ijassa-122	121	25	innovation	innovation	NOUN
ijassa-122	121	26	outliers	outlier	NOUN
ijassa-122	121	27	in	in	ADP
ijassa-122	121	28	time	time	NOUN
ijassa-122	121	29	series	series	PROPN
ijassa-122	121	30	”	"	PUNCT
ijassa-122	121	31	,	,	PUNCT
ijassa-122	121	32	computational	computational	ADJ
ijassa-122	121	33	statistics	statistic	NOUN
ijassa-122	121	34	&	&	CCONJ
ijassa-122	121	35	data	datum	NOUN
ijassa-122	121	36	analysis	analysis	NOUN
ijassa-122	121	37	,	,	PUNCT
ijassa-122	121	38	vol.37	vol.37	NOUN
ijassa-122	121	39	,	,	PUNCT
ijassa-122	121	40	pp.1	pp.1	NOUN
ijassa-122	121	41	-	-	PUNCT
ijassa-122	121	42	12	12	NUM
ijassa-122	121	43	.	.	PUNCT
ijassa-122	122	1	advances	advance	NOUN
ijassa-122	122	2	in	in	ADP
ijassa-122	122	3	systems	system	NOUN
ijassa-122	122	4	science	science	NOUN
ijassa-122	122	5	and	and	CCONJ
ijassa-122	122	6	applications	application	NOUN
ijassa-122	122	7	(	(	PUNCT
ijassa-122	122	8	2012	2012	NUM
ijassa-122	122	9	)	)	PUNCT
ijassa-122	122	10	vol.12	vol.12	NOUN
ijassa-122	122	11	no.4	no.4	PROPN
ijassa-122	122	12	405	405	NUM
ijassa-122	122	13	[	[	X
ijassa-122	122	14	2	2	X
ijassa-122	122	15	]	]	PUNCT
ijassa-122	122	16	peña	peña	PROPN
ijassa-122	122	17	d	d	PROPN
ijassa-122	122	18	,	,	PUNCT
ijassa-122	122	19	sánchez	sánchez	PROPN
ijassa-122	122	20	i.	i.	NOUN
ijassa-122	122	21	(	(	PUNCT
ijassa-122	122	22	2005	2005	NUM
ijassa-122	122	23	)	)	PUNCT
ijassa-122	122	24	,	,	PUNCT
ijassa-122	122	25	“	"	PUNCT
ijassa-122	122	26	multifold	multifold	VERB
ijassa-122	122	27	predictive	predictive	ADJ
ijassa-122	122	28	validation	validation	NOUN
ijassa-122	122	29	in	in	ADP
ijassa-122	122	30	armax	armax	ADJ
ijassa-122	122	31	time	time	NOUN
ijassa-122	122	32	series	series	PROPN
ijassa-122	122	33	models	model	NOUN
ijassa-122	122	34	”	"	PUNCT
ijassa-122	122	35	,	,	PUNCT
ijassa-122	122	36	journal	journal	NOUN
ijassa-122	122	37	of	of	ADP
ijassa-122	122	38	the	the	DET
ijassa-122	122	39	american	american	PROPN
ijassa-122	122	40	statistical	statistical	PROPN
ijassa-122	122	41	association	association	PROPN
ijassa-122	122	42	,	,	PUNCT
ijassa-122	122	43	vol.100	vol.100	PROPN
ijassa-122	122	44	,	,	PUNCT
ijassa-122	122	45	pp.135	pp.135	NOUN
ijassa-122	122	46	-	-	X
ijassa-122	122	47	146	146	NUM
ijassa-122	122	48	.	.	PUNCT
ijassa-122	123	1	[	[	X
ijassa-122	123	2	3	3	X
ijassa-122	123	3	]	]	X
ijassa-122	123	4	chen	chen	PROPN
ijassa-122	123	5	p	p	PROPN
ijassa-122	123	6	,	,	PUNCT
ijassa-122	123	7	li	li	PROPN
ijassa-122	123	8	l	l	PROPN
ijassa-122	123	9	,	,	PUNCT
ijassa-122	123	10	liu	liu	PROPN
ijassa-122	123	11	y	y	PROPN
ijassa-122	123	12	and	and	CCONJ
ijassa-122	123	13	lin	lin	PROPN
ijassa-122	123	14	j.g	j.g	PROPN
ijassa-122	123	15	.	.	PROPN
ijassa-122	123	16	(	(	PUNCT
ijassa-122	123	17	2010	2010	NUM
ijassa-122	123	18	)	)	PUNCT
ijassa-122	123	19	,	,	PUNCT
ijassa-122	123	20	“	"	PUNCT
ijassa-122	123	21	detection	detection	NOUN
ijassa-122	123	22	of	of	ADP
ijassa-122	123	23	outliers	outlier	NOUN
ijassa-122	123	24	and	and	CCONJ
ijassa-122	123	25	patches	patch	NOUN
ijassa-122	123	26	in	in	ADP
ijassa-122	123	27	bilinear	bilinear	PROPN
ijassa-122	123	28	time	time	NOUN
ijassa-122	123	29	series	series	PROPN
ijassa-122	123	30	models	model	NOUN
ijassa-122	123	31	”	"	PUNCT
ijassa-122	123	32	,	,	PUNCT
ijassa-122	123	33	mathematical	mathematical	ADJ
ijassa-122	123	34	problems	problem	NOUN
ijassa-122	123	35	in	in	ADP
ijassa-122	123	36	engineering	engineering	NOUN
ijassa-122	123	37	,	,	PUNCT
ijassa-122	123	38	vol.2010	vol.2010	NOUN
ijassa-122	123	39	,	,	PUNCT
ijassa-122	123	40	pp.1	pp.1	NOUN
ijassa-122	123	41	-	-	PUNCT
ijassa-122	123	42	10	10	NUM
ijassa-122	123	43	.	.	PUNCT
ijassa-122	124	1	[	[	X
ijassa-122	124	2	4	4	X
ijassa-122	124	3	]	]	X
ijassa-122	124	4	chen	chen	PROPN
ijassa-122	124	5	p	p	PROPN
ijassa-122	124	6	,	,	PUNCT
ijassa-122	124	7	yang	yang	PROPN
ijassa-122	124	8	j	j	PROPN
ijassa-122	124	9	,	,	PUNCT
ijassa-122	124	10	li	li	PROPN
ijassa-122	124	11	l.y	l.y	PROPN
ijassa-122	124	12	.	.	PROPN
ijassa-122	124	13	(	(	PUNCT
ijassa-122	124	14	2013	2013	NUM
ijassa-122	124	15	)	)	PUNCT
ijassa-122	124	16	,	,	PUNCT
ijassa-122	124	17	“	"	PUNCT
ijassa-122	124	18	synthetic	synthetic	ADJ
ijassa-122	124	19	detection	detection	NOUN
ijassa-122	124	20	of	of	ADP
ijassa-122	124	21	change	change	NOUN
ijassa-122	124	22	point	point	NOUN
ijassa-122	124	23	and	and	CCONJ
ijassa-122	124	24	outliers	outlier	NOUN
ijassa-122	124	25	in	in	ADP
ijassa-122	124	26	bilinear	bilinear	PROPN
ijassa-122	124	27	time	time	NOUN
ijassa-122	124	28	series	series	PROPN
ijassa-122	124	29	models	model	NOUN
ijassa-122	124	30	”	"	PUNCT
ijassa-122	124	31	,	,	PUNCT
ijassa-122	124	32	international	international	ADJ
ijassa-122	124	33	journal	journal	NOUN
ijassa-122	124	34	of	of	ADP
ijassa-122	124	35	systems	system	NOUN
ijassa-122	124	36	science	science	NOUN
ijassa-122	124	37	,	,	PUNCT
ijassa-122	124	38	http://dx.doi.org/10.1080/00207721.2013.777983	http://dx.doi.org/10.1080/00207721.2013.777983	NOUN
ijassa-122	124	39	.	.	PUNCT
ijassa-122	125	1	(	(	PUNCT
ijassa-122	125	2	in	in	ADP
ijassa-122	125	3	press	press	NOUN
ijassa-122	125	4	)	)	PUNCT
ijassa-122	126	1	[	[	X
ijassa-122	126	2	5	5	X
ijassa-122	126	3	]	]	X
ijassa-122	126	4	huang	huang	PROPN
ijassa-122	126	5	l	l	PROPN
ijassa-122	126	6	,	,	PUNCT
ijassa-122	126	7	pang	pang	PROPN
ijassa-122	126	8	w	w	PROPN
ijassa-122	126	9	,	,	PUNCT
ijassa-122	126	10	wang	wang	PROPN
ijassa-122	126	11	k.p	k.p	PROPN
ijassa-122	126	12	,	,	PUNCT
ijassa-122	126	13	zhou	zhou	PROPN
ijassa-122	126	14	c.g	c.g	PROPN
ijassa-122	126	15	,	,	PUNCT
ijassa-122	126	16	xiao	xiao	PROPN
ijassa-122	126	17	y.	y.	PROPN
ijassa-122	126	18	(	(	PUNCT
ijassa-122	126	19	2004	2004	NUM
ijassa-122	126	20	)	)	PUNCT
ijassa-122	126	21	,	,	PUNCT
ijassa-122	126	22	“	"	PUNCT
ijassa-122	126	23	improved	improve	VERB
ijassa-122	126	24	genetic	genetic	ADJ
ijassa-122	126	25	algorithm	algorithm	NOUN
ijassa-122	126	26	for	for	ADP
ijassa-122	126	27	vehicle	vehicle	NOUN
ijassa-122	126	28	routing	routing	NOUN
ijassa-122	126	29	problem	problem	NOUN
ijassa-122	126	30	with	with	ADP
ijassa-122	126	31	time	time	NOUN
ijassa-122	126	32	windows	window	NOUN
ijassa-122	126	33	”	"	PUNCT
ijassa-122	126	34	,	,	PUNCT
ijassa-122	126	35	advances	advance	NOUN
ijassa-122	126	36	in	in	ADP
ijassa-122	126	37	systems	system	NOUN
ijassa-122	126	38	science	science	NOUN
ijassa-122	126	39	and	and	CCONJ
ijassa-122	126	40	applications	application	NOUN
ijassa-122	126	41	,	,	PUNCT
ijassa-122	126	42	vol.4	vol.4	PROPN
ijassa-122	126	43	,	,	PUNCT
ijassa-122	126	44	pp.118	pp.118	PROPN
ijassa-122	126	45	-	-	PROPN
ijassa-122	126	46	124	124	NUM
ijassa-122	126	47	.	.	PUNCT
ijassa-122	127	1	[	[	X
ijassa-122	127	2	6	6	NUM
ijassa-122	127	3	]	]	PUNCT
ijassa-122	127	4	box	box	PROPN
ijassa-122	127	5	g.e.p	g.e.p	PROPN
ijassa-122	127	6	,	,	PUNCT
ijassa-122	127	7	jenkins	jenkins	PROPN
ijassa-122	127	8	g.m	g.m	PROPN
ijassa-122	127	9	,	,	PUNCT
ijassa-122	127	10	reinsel	reinsel	PROPN
ijassa-122	127	11	g.c	g.c	PROPN
ijassa-122	127	12	.	.	PROPN
ijassa-122	127	13	(	(	PUNCT
ijassa-122	127	14	1994	1994	NUM
ijassa-122	127	15	)	)	PUNCT
ijassa-122	127	16	,	,	PUNCT
ijassa-122	127	17	time	time	NOUN
ijassa-122	127	18	series	series	PROPN
ijassa-122	127	19	analysis	analysis	NOUN
ijassa-122	127	20	:	:	PUNCT
ijassa-122	127	21	forecasting	forecasting	NOUN
ijassa-122	127	22	and	and	CCONJ
ijassa-122	127	23	control	control	NOUN
ijassa-122	127	24	,	,	PUNCT
ijassa-122	127	25	third	third	ADJ
ijassa-122	127	26	edition	edition	NOUN
ijassa-122	127	27	,	,	PUNCT
ijassa-122	127	28	prentice	prentice	NOUN
ijassa-122	127	29	-	-	PUNCT
ijassa-122	127	30	hall	hall	NOUN
ijassa-122	127	31	,	,	PUNCT
ijassa-122	127	32	englewood	englewood	PROPN
ijassa-122	127	33	cliffs	cliffs	PROPN
ijassa-122	127	34	,	,	PUNCT
ijassa-122	127	35	nj	nj	PROPN
ijassa-122	127	36	.	.	PUNCT
ijassa-122	128	1	corresponding	correspond	VERB
ijassa-122	128	2	author	author	NOUN
ijassa-122	128	3	author	author	NOUN
ijassa-122	128	4	can	can	AUX
ijassa-122	128	5	be	be	AUX
ijassa-122	128	6	contacted	contact	VERB
ijassa-122	128	7	at	at	ADP
ijassa-122	128	8	cp18@263.net.cn	cp18@263.net.cn	NOUN