Asian Journal of Economics and Empirical Research Vol. 4, No. 1, 32-47, 2017 ISSN(E) 2409-2622 / ISSN(P) 2518-010X DOI: 10.20448/journal.501.2017.41.32.48 32 The Enterprise Risk Management of Foreign Exchange Exposures: Evidence from Taiwanese Hospitality Industry Hsiao, Chiu-Ming1  Zhang, Wei-Fang2 ( Corresponding Author) Chiu, Chi-Chang3 Huang, Jung-Chang4 Huang, Yu-Ling5 1,2,3,4,5Department of Finance, National Yunlin University of Science & Technology, Taiwan Abstract For this paper, I use the ARIMA model to study the relationship between business performance and exchange rate fluctuations. Through this model, the empirical results shows that the influences of foreign exchange rate fluctuations on the tourist hotel business performance are significant and different across currencies and firms. Furthermore, according to the framework of Kim (2013) we employ the modern portfolio theory proposed by Markowitz (1952) to give an optimal foreign exchange allocation for each tourist hotel company's financial decision-makers, which will avoid the risk of exchange rate fluctuations expose and reduce losses due to the fluctuations of exchange rates, and complete the construction of enterprise risk management system (ERM). Keywords: Foreign exchange exposures, Modern portfolio theory, Enterprise risk management. Citation | Hsiao, Chiu-Ming; Zhang, Wei-Fang; Chiu, Chi-Chang; Huang, Jung-Chang; Huang, Yu-Ling (2017). The Enterprise Risk Management of Foreign Exchange Exposures: Evidence from Taiwanese Hospitality Industry. Asian Journal of Economics and Empirical Research, 4(1): 32-48. History: Received: 6 April 2017 Revised: 14 September 2017 Accepted: 18 September 2017 Published: 23 September 2017 Licensed: This work is licensed under a Creative Commons Attribution 3.0 License Publisher: Asian Online Journal Publishing Group Contribution/Acknowledgement: All authors contributed to the conception and design of the study. Funding: This study received no specific financial support. Competing Interests: The authors declare that they have no conflict of interests. Transparency: The authors confirm that the manuscript is an honest, accurate, and transparent account of the study was reported; that no vital features of the study have been omitted; and that any discrepancies from the study as planned have been explained. Ethical: This study follows all ethical practices during writing. Contents 1. Introduction ...................................................................................................................................................................................... 33 2. Methodologies .................................................................................................................................................................................. 34 3. Data .................................................................................................................................................................................................... 37 4. Conclusions ....................................................................................................................................................................................... 47 References .............................................................................................................................................................................................. 47 http://creativecommons.org/licenses/by/3.0/ http://creativecommons.org/licenses/by/3.0/ https://orcid.org/orcid-search/quick-search?searchQuery=Hsiao, Chiu-Ming https://orcid.org/orcid-search/quick-search?searchQuery=Zhang, Wei-Fang https://orcid.org/orcid-search/quick-search?searchQuery=Chiu, Chi-Chang https://orcid.org/orcid-search/quick-search?searchQuery=Huang, Jung-Chang https://orcid.org/orcid-search/quick-search?searchQuery=Huang, Yu-Ling https://orcid.org/orcid-search/quick-search?searchQuery=Hsiao, Chiu-Ming https://orcid.org/orcid-search/quick-search?searchQuery=Zhang, Wei-Fang https://orcid.org/orcid-search/quick-search?searchQuery=Chiu, Chi-Chang https://orcid.org/orcid-search/quick-search?searchQuery=Huang, Jung-Chang https://orcid.org/orcid-search/quick-search?searchQuery=Huang, Yu-Ling https://orcid.org/orcid-search/quick-search?searchQuery=Hsiao, Chiu-Ming https://orcid.org/orcid-search/quick-search?searchQuery=Zhang, Wei-Fang https://orcid.org/orcid-search/quick-search?searchQuery=Chiu, Chi-Chang https://orcid.org/orcid-search/quick-search?searchQuery=Huang, Jung-Chang https://orcid.org/orcid-search/quick-search?searchQuery=Huang, Yu-Ling https://orcid.org/orcid-search/quick-search?searchQuery=Hsiao, Chiu-Ming https://orcid.org/orcid-search/quick-search?searchQuery=Zhang, Wei-Fang https://orcid.org/orcid-search/quick-search?searchQuery=Chiu, Chi-Chang https://orcid.org/orcid-search/quick-search?searchQuery=Huang, Jung-Chang https://orcid.org/orcid-search/quick-search?searchQuery=Huang, Yu-Ling https://orcid.org/orcid-search/quick-search?searchQuery=Hsiao, Chiu-Ming https://orcid.org/orcid-search/quick-search?searchQuery=Zhang, Wei-Fang https://orcid.org/orcid-search/quick-search?searchQuery=Chiu, Chi-Chang https://orcid.org/orcid-search/quick-search?searchQuery=Huang, Jung-Chang https://orcid.org/orcid-search/quick-search?searchQuery=Huang, Yu-Ling https://orcid.org/orcid-search/quick-search?searchQuery=Hsiao, Chiu-Ming https://orcid.org/orcid-search/quick-search?searchQuery=Zhang, Wei-Fang https://orcid.org/orcid-search/quick-search?searchQuery=Chiu, Chi-Chang https://orcid.org/orcid-search/quick-search?searchQuery=Huang, Jung-Chang https://orcid.org/orcid-search/quick-search?searchQuery=Huang, Yu-Ling Asian Journal of Economics and Empirical Research, 2017, 4(1): 32-48 33 1. Introduction Tourism industry has named the no-smokestack industry. The revenue generates from the tourism industry will increase as the growth of inbounds and the time period that they stay. As the economic viewpoint, tourism will create value from catering, hotels, aviation, transport and many other related industries. It also helps to revitalize the tourism industry association actives, the economic benefits, not only to create a tourism value, it also increases consumption and further boosts the economy, increases employment opportunities. In 2013, Japanese Prime Minister Shinzo Abe implemented a policy combining fiscal expansion, “quantitative easing”, and structural reform in the hope of revitalizing Japan„s domestic economy. Indeed, this is so-called ― Abenomics--- results in a significant growth in Japan„s domestic economy. Accordingly, the impact of exchange rates on some industry becomes even more obvious and important, especially in the tourism industry while the Japanese yen is depreciated in order to stimulate the economy fast. The impact of exchange rates on the part of the industry becomes even more obvious and important, especially in the tourism industry by the Japanese yen depreciated under the influence of the economy back to temperature very fast. Implementing quantitative easing policy that caused the depreciation of Japanese yen increases Japan„s foreign trade and also successfully lead the economy back to situation. Surprisingly, the tourism industry has gained the most benefits of all. This paper, therefore, wants to study the case and examine if the situation could as well apply to the tourism industry in Taiwan. Oh (2005) addressed the causal relations between tourism growth and economic expansion for the Korean economy. He employed the Granger causality test and found that the Korean tourism industry is economic-driven. Kim et al. (2006) examined the relationship between tourism expansion and economic development in Taiwan. They found a bi-directional causality between them. In other words, in Taiwan, tourism expansion and economic development reinforce each other. Min (2013) used panel data approach to test the tourism-led economic growth hypothesis. He found that the tourism-led growth hypothesis is more strongly supported when the time-specific effects are eliminated, which will cause a biased estimate in the Granger causality test. According to the data of the World Tourism Organization in April 2015 announcement, the number of international tourist visited in Taiwan in 2012 was estimated about 9.9 million, ranked as the world‟s 31 and created revenues about $ 14.7 billion. In 2014, Taiwan inbound tourists grew 23.6%, ranking the 2nd place of the world‟s top 50 tourist destinations, only less than the Japan‟s growth rate of 29.4%. Tourism revenue has growth 18.9%, ranking the 4th place in the world‟s top 50 tourism revenue areas. Gradually, Taiwan‟s tourism has been recognized considerable potential. Chen and Zan (2009) have showed that the tourism industry greatly contributed to the Taiwanese economy that is, Taiwan is tourism-led economic. Taiwan authority has opened the Chinese group tourists to Taiwan since the summer of 2008. In order to push up the number of tourists, Taiwan government implements many projects to develop the tourism industry, such as Doubling Tourist Arrivals Plan (DTAP) introduced in 2002 and „„Challenge 2008‟‟, Taiwan‟s 2015-2018 Tourism Action Plan, Mid-term Plan for Construction of Major Scenic Sites (2012-2015), Project Vanguard for Excellence in Tourism, Tour Taiwan and Experience the Centennial, etc. According to Taiwan Tourism Bureau, these plans are proposed to deepen of the “Time for Taiwan” core promotional program, and, to use “quality, uniqueness, intelligence, and sustainability” as strategies toward the goals of “development of international tourism, enhancement of domestic travel quality, and increased foreign-exchange revenues” to bring Taiwan‟s new tourism allure to the attention of the world1. On the other hand, Portnov and Li (2013) suggested that in order to achieve a greater stability in the number of inbound tourist arrivals, Taiwan should diversify sources of their inbound tourism, by giving priority to neighboring countries with relatively larger, more productive, and more steadily growing economies, such as China, Malaysia, Thailand, or the other emerging countries in South-Eastern Asia. According to the Taiwan Tourism Bureau, the inbound number of tourists was 2,624,037 in 2000, 9,910,204 in 2014, growing about 3.8 times and over 10 million in the end of 2015. This tendency shows the visibility and attractiveness of international tourism in traveling to Taiwan, which significantly increase the number of attentions. Moreover, Taiwan‟s foreign exchange earnings generated by tourism was from $3,738 million in 2000 to $14,615 million in 2014 and its share in total GDP also reach 2.76% from 1.13%. It shows that Taiwan tourism industry earns a huge of foreign exchange earnings. Thus, the fluctuations in exchange rates for Taiwan‟s tourism industry is also an important factor for Taiwan‟s overall economic development. The recent ten-year annual revenues generate from tourism, foreign exchange and domestic tourism are shown in the Figure 1. The highest line is the tourism revenue (in red) which grows rapidly in 2009 due to the effect of opening of Chinese tourists to visit Taiwan. And the lowest line is the domestic tourism revenue (in purple) which attains the maximum (331 billion of NT dollars) in 2011 and declines in the following years. The foreign exchange earnings (in green) smoothly increases in years. 1 http://admin.taiwan.net.tw/public/public_en.aspx?no=6. http://admin.taiwan.net.tw/public/public_en.aspx?no=6 Asian Journal of Economics and Empirical Research, 2017, 4(1): 32-48 34 Figure-1. Taiwan annual revenues generate from tourism, foreign exchange, and domestic tourism Source: Tourism Bureau, M.O.T.C., Republic of China (Taiwan) Taiwan„s tourism revenues have increased in recent years, the tourism industry plays an important role in the tourism industry, resulting in a huge source of foreign exchange earnings. Among the tourists, the number of Chinese tourists accounted for the largest cases, followed by Japanese, European and the United States. Bilateral trade between Taiwan and China, Japan, Europe and the United States, respectively is not only very close, but also represents the effect of the changes in exchange rates. The number of tourists traveling to Taiwan contributes the foreign exchange earnings. On the other hand, Pritamani et al. (2005) divided the U.S. companies into five categories and found that neither exporters nor multinational firms were the most affected by changes in exchange rates. The firms that suffered most from exchange rate fluctuations were wholly domestic U.S. companies facing foreign competition. Taiwan‟s hotel industry has the same situation. Based on the above point of view, we mainly discuss Taiwan‟s hotel industry for exposure to foreign exchange fluctuations and corporate risk management. Through our study, it suggests the hedging strategies to the decision-makers of firms and then to enhance Taiwan‟s hotel industry‟s risk management. The structure of our study is: Methodologies will be discussed in Section II; data collection and their statistical descriptions are in the Section 3. The empirical results and their analysis are shown in the Section 4. The last section is our conclusion. 2. Methodologies A. Modern Portfolio Theory(MPT) Modern portfolio theory is proposed by Markowitz in 1952. In the paper, Probability Theory and Linear Algebra method are applied to investigate the correlation between the securities. It put forward the possibility to diversify the main investment risks for this theory is that the risks associated with some other securities regardless of the dispersion of individual investment targets can reduce the risk. In this way, individual company information becomes less important. The theory is mainly to solve the investor‟s risk-reward problem and to form a rational combination of its own funds in order to maximize the proceeds. According to the Markowitz‟s framework, there is a certain special relationship between investment risk and return of a portfolio of financial assets. His assumptions: 1. Assume the market is efficient, investors can learn more of the benefits and risks of financial market changes and their causes. 2. Suppose investors are risk averse and are willing to get a higher rate of return if they must bear a greater risk to get a higher expected return as compensation. Risk is the variability of yields as measured by standard deviation. 3. Investors‟ choices are based on the expected returns and standard deviations of selected financial assets portfolio. They select portfolios with higher yields or lower risk. 4. The incomes between various financial assets are correlated with the correlation coefficient between each financial asset, it is possible to choose the lowest risk of the portfolio. And an efficient portfolio, it should be subject to the following conditions: under certain risk (standard deviation), this combination of securities has the highest average reward; and in certain average reward, it has the lowest degree of risk (standard deviation). Therefore, the portfolio should be on the curve of efficient frontier. According to Huang and Litzenberger (1988); Elton et al. (2007) suppose an economy which there are n risky assets with its return and standard deviation iR and i , ni , , , 21 , respectively. Moreover, the covariance between any two assets is  jiji RRCov , , ， nji , , , 21 . If we denote the portfolio weight on each assets in the portfolio to be iw , ni , , , 21 , then the expected return of the portfolio is RwP  , where Asian Journal of Economics and Empirical Research, 2017, 4(1): 32-48 35   nRRRR , , , 21 and   nwwww , , , 21 。And the variance of the portfolio is wwP 2 , where     nnjiRVar   ,  is the variance-covariance matrix. Hence, in the framework of Markowitz (1952) and Kim (2013) we have to minimize the degree of risk of the portfolio under a pre-specified return, 0 , and budget constrain. Namely, wwP wi  2 1 2 2 min  (1)             niw wwJ Rw ts i n i in P , , , , 2110 1.. 1 0 (2) where,   n nJ    111 , , ,  . Using the Lagrange Multipliers method, the above problem can be transformed as follows:             wJRwwww n wi 1 2 1 20121min 21   , , , ,  . (3) Hence, the F.O.C.is                                 01 0 0 2 0 1 21 Pn P nP wJ Rw JRw w         nP JRw   1 2 1 1 *  . (4) And then we have,               11 2 1 1 0 1 2 1 1 nnn n JJRJ RJRR                  02 0 2 02 0 1       D B D A BAC BA D B D C BAC BC , where, RRA  1 , nn JRRJB     11 , nn JJC    1 , and 2BACD  . Such that, the optimal wealth allocation portfolio is nP JRw   1 2 1 1 *  nJ D B D A R D B D C               1 0 1 0  . (5) The properties of this portfolio are 1. RJ D B D A RR D B D C Rw nPP                   1 0 1 0 *  .000 2 00                 D D D BAC B D B D A A D B D C 2. **2 PPP ww                          nJ D B D A R D B D C 1 0 1 0                      nJ D B D A R D B D C 1 0 1 0  nn JJ D B D A RR D B D C                 1 2 0 1 2 0  nJR D B D A D B D C              1 002                           00 2 0 2 0 2  D B D A D B D C B D B D A C D B D C A   CC B D C ABC D 1 2 1 2 00 2 0         (6) Such that, C P 12  and the equality holds when C B 0 . Asian Journal of Economics and Empirical Research, 2017, 4(1): 32-48 36 Next, considering a riskless asset can be invested, and then the pre-described model will be rewritten as follows: wwP wi  2 1 2 2 min  (7)   01..   RwrJwts fnP , (8) where, fr is the return of the riskless assets. Again, by using the Lagrange Multipliers method, we have to solve the following problem:     fn w rJwRwwww i  1 2 1 021min   , , ,  . (9) Thus, the F.O.C. is         01 0 0 fn fnw rJwRw rJRw        H r JrRw f nfP    01*  , (10) where,     21 2 ffnfnf rCrBAJrRJrRH     . (11) The properties of this portfolio are: 1. fnPPP rJwRw           1*   fnP f nf rJwR H r JrR              1 01          nnf ff nfnf f JJrR H rr JrRJrR H r           1010  fnP rJw          *1 . (12) 2. **2 PPP ww                          H r JrR H r JrR f nf f nf 0101      nfnf f JrRJrR H r             1 2 0                  nnfnf f JJrJRrRR H r 121 2 0 2    H r f 2 0    (13) Hence, H r f P   0  , that is, Pf Hr   0 (14) B. Autoregression Integrated Moving Average Models,  qdpARIMA ,, In Witt and Witt (1992;1995) they use many econometric models to investigate the topics of tourism industries. Empirically, they suggested that the autoregression and moving average models can be implemented to forecast the performance of tourism industries. Here, we want to investigate the effects of the fluctuations of foreign exchange on the performance of hotel industry. According to Bodie et al. (2002) we can use the ROA or ROE, reported in the annual financial statements, to be the measures of the corporate‟s performance. There are at least two reasons for applying ROA/ROE to proxy the firm‟s performance. First, since ROA is the return of corporate‟s total assets, which is defined by the product of profit margin and total asset turnover, so it tells us how effectively a firm uses its assets to generate profits. Therefore, a well-performed firm will have a higher ROA. Second, the definition of ROE is the net profit over the average equity, so that by the DuPont equation, we have RatioEquity Asset TurnoverAsset MarginProfit Net ROE . (15) Hence, it tells us how efficiently a company is operated. It also provides insights into the firm‟s use of assets via turnover. That is, a well-performed firm also has a higher ROE. As a result, in our study, we will apply these two measures to be the proxies of the firm‟s performance and investigate the magnitude of the effects of foreign exchange rate‟s fluctuations. First, Dumas (1978); Adler and Bernard (1980) and Hodder (1982) implement the change of foreign exchange rates into the regression models to study the U.S. multinational firm‟s values. And Jorion (1990) followed their studies and found that the stock returns of U.S. multinational firm are significantly positively correlated to the volatility of the U.S. dollar. Moreover, Bodnar and William (1993) studied the different effects of the fluctuations of foreign exchange rates on the different industries in U.S., Canada and Japan. And in Hamid et al. (2013) discussed Asian Journal of Economics and Empirical Research, 2017, 4(1): 32-48 37 the public relations agency (PRA) in the People‟s Republic of China (PRC) by using importance-performance analysis (IPA). Moreover, He and Ng (1998) investigated Japan 171 multinational firms there are about 25% firm‟s stock returns significantly positively correlated to the foreign exchange exposures, themselves. And the effects are increasing as firm‟s size increases. Moreover, in Fama and Kenneth (1993;1995) they formed six portfolios of the stocks listed on NYSE, AMX, and NASDAQ Stock Market by the firm‟s size and found that firm‟s size and BE/ME proxy for sensitivity to risk factors that capture strong common variation in stock returns and will help to explain the average returns and then firm‟s profitability. And Morelli (2007) found the same effects of firm‟s size on the UK listed firms‟ stock returns. Their results showed that the media personnel and travel agents/tour operators were basically satisfied with the PRA‟s performance, although there is still room for improvement. On the other hand, Maloney (1990) paid attention on the Australian mining firms. He indicated that the fluctuations of the exchange rates between Australia dollars against to the major currencies will affect the firm‟s profit. So he suggested that firm should find some strategies to manage the positions of foreign currencies in order to avoid the losses caused by the fluctuations of exchange rates and then reduce the firm‟s performance. Bailey et al. (1992) and Kim (2013) suggested that multinational enterprise may use the framework of the Modern Portfolio Theory to form their own foreign exchange risk management strategies and to reduce the effect of foreign exchange exposures. Here, we apply the framework of Kim to investigate the effects of foreign exchange exposures on the performance of Taiwan hospitality industry and try to propose some hedging strategies and strengthen their corporate risk management. Therefore, in our regression models, we will impose the changes of exchange rates of several currencies to study the effects of the fluctuations of exchange rates on the performance of Taiwan hospitality firms. Our autoregression moving average model is given as follows:    p k ktikitmiiti ePerformancRMRFePerformanc 1 , , , ,       q s stistii n j tjji aSizeFX 01 , , , ,  , (16)     iTqpqpt , , , , , 2max1max  , Ni , , , 21 . Where, tiePerformanc , represents the ith firm‟s performance in the tth quarter, and ktiePerformanc  , is its kth lagged variable. In Sharpe (1964) he defined that tRMRF is the market portfolio‟s excess return in the t-th quarter, i.e., ftt rRmRMRF  , tRm is the market portfolio‟s return and fr is the rate of return of riskless asset. Furthermore, as indicated in Smithson and Simkins (2005) although the management of interest rate and foreign exchange rate risks does indeed add value, the effect is larger than would be expected. Such that, let tjFX ,  be the percentage change of exchange rate of the jth currency in the tth quarter, which is defined by 100 1 1      tj tjtj tj E EE FX , , , , , (17) where tjE , is the closed price in the end of the quarter in terms of direct quotation. And tiSize , is the size of the ith firm in the tth quarter which is defined as  titi CapSize , , ln , and tiCap , is the capitalization of the firm in the tth quarter. tia , are the white noises. 3. Data This paper selected 12 hospitality companies listed on Taiwan Stock Exchange (TWSE), and downloaded their quarterly ROA, ROE and capitalization from Taiwan Economic Journal (TEJ). They are Hotel Holiday Garden (2702), The Ambassador Hotel Ltd. (2704), The Leofoo Development Co., Ltd. (2705), First Hotel Company Ltd. (2706), Formosa International Hotels Corporation (2707), Farglory Hotel Co., Ltd. (2712), Pleasant Hotels International Inc. (2718), Chateau International Development Co., Ltd. (2722), FX Hotels Group Inc. (2724-F), Janfusun Fancyworld Corp. (5701), The Landis Taipei Hotel Co., Ltd. (5703), and Hotel Royal Chihpen (5704). Period is from 2000Q1 to 2015Q3 and sum to 489 firm-quarters. Table 1 shows the descriptive statistics of the firm‟s ROA and ROE, respectively. Table-1. (A). Descriptive statistics of ROA. ROA (%) Obs. Mean Std. dev. Max Min Median 2702 HG 63 0.661 0.820 2.94 -1.43 0.740 2704 AMBH 63 0.641 0.565 1.47 -1.29 0.740 2705 Leofoo 32 -0.136 1.298 5.40 -5.03 -0.225 2706 First Hotel 32 1.398 1.105 7.28 0.54 1.160 2707 GFRT 63 4.392 1.195 7.55 1.27 4.360 2712 FGH 11 1.383 1.244 3.68 -0.12 0.870 2718 PH 25 0.944 0.873 2.40 -1.20 0.840 2722 Chateau 21 2.732 2.632 8.38 -0.60 2.380 2724 FX Hotels 21 1.179 1.770 5.74 -2.45 1.550 5701 JFS 32 -1.462 1.341 1.73 -5.76 -1.470 5703 Landis Taipei 63 0.419 1.724 3.20 -8.97 0.740 5704 Chihpen Royal 63 1.040 1.381 3.62 -3.49 1.210 Source: Taiwan Economic Journal (TEJ). Asian Journal of Economics and Empirical Research, 2017, 4(1): 32-48 38 Table-1(B). Descriptive statistics of ROE. ROE (%) Obs. Mean Std. dev. Max Min Median 2702 HG 63 0.804 1.276 3.77 -2.97 0.940 2704 AMBH 63 0.704 0.956 2.18 -2.96 0.880 2705 Leofoo 32 -0.415 2.565 11.04 -9.90 -0.695 2706 First Hotel 32 1.668 1.339 8.71 0.67 1.390 2707 GFRT 63 6.645 2.205 11.13 1.50 6.700 2712 FGH 11 1.794 1.782 4.81 -0.44 1.100 2718 PH 25 1.220 1.109 2.99 -1.62 1.160 2722 Chateau 21 3.313 3.335 11.24 -0.68 3.060 2724 FX Hotels 21 1.418 3.958 6.96 -8.83 2.210 5701 JFS 32 -3.217 2.462 2.47 -11.24 -3.300 5703 Landis Taipei 63 0.612 2.338 4.48 -11.76 1.040 5704 Chihpen Royal 63 1.192 1.581 4.19 -3.83 1.300 Source: Taiwan Economic Journal (TEJ). In Table 1, we may find that the Formosa International Hotels Corporation (2707) has the highest ROA and ROE, however, Janfusun Fancyworld Corp. (5701) has the lowest ROA and ROE. And except of Janfusun Fancyworld Corp. and the Leofoo Development Co., Ltd. (2705), the others are well-performed since they all have a positive average ROA or ROE. Moreover, the Ambassador Hotel Ltd. (2704) has the lowest volatility of ROA and ROE. On the other hand, Chateau International Development Co., Ltd. (2722) and the FX Hotels Group Inc. (2724-F) have the highest volatility of ROA and ROE, respectively. It may result from the shortest listing data of these two companies. Next, we collect the foreign exchange rates from the website of the Central Bank of Taiwan. The data period is from 2000 to 2015. And then calculate the quarterly and monthly percentage change of exchange rates for the currencies against to the NT dollars (NTD) according to the Equation (13). Table 2 shows the descriptive statistics of the monthly change of foreign exchange rates. Table-2. Descriptive statistics of the monthly change of exchange rates. Monthly Change (%) Mean Std. dev. Max Min Median CV USD 0.0280 1.1865 3.3313 -3.5798 0.0232 42.3750 JPY -0.0353 2.3233 8.8309 -6.0498 -0.2280 -65.8159 GBP 0.0135 1.9269 5.7608 -8.0517 0.0901 142.7333 CNY 0.1664 1.1449 3.0715 -3.4686 0.1749 6.8804 EUR 0.1016 2.2251 7.1016 -5.2660 0.2209 21.9006 HKD 0.0294 1.1841 3.3179 -3.5798 0.0205 40.2755 KRW 0.0170 1.8737 6.7195 -12.3726 0.2496 110.2176 CAD 0.0940 1.8282 5.0170 -8.6264 0.0351 19.4489 SGD 0.1155 0.9621 3.0253 -3.4506 0.1384 8.3299 AUD 0.1074 2.6040 6.9746 -13.5568 0.2725 24.2458 IDR -0.3005 3.1096 20.7023 -13.5534 -0.2083 -10.3481 THB 0.0602 1.2869 3.6022 -4.5300 0.0806 21.3771 MYR -0.0366 1.2266 3.0689 -4.0603 -0.0825 -33.5137 PHP -0.0412 1.5530 4.3555 -4.5580 -0.1633 -37.6942 Source: Central Bank of Taiwan. http://www.cbc.gov.tw/content.asp?mp=1&CuItem=36599. In Table 2, the lowest percentage change (0.96%) of the exchange rate is the Singapore dollar exchange rate against to NT dollar and has the highest percentage change (3.11%) of the exchange rate is the Indonesian rupiah exchange rate against to NT dollar. Since Indonesian rupiah has a maximum appreciation (20.70%) and minimum depreciation (13.56%) against to NT dollar. Moreover, the coefficient of variation, a nominal measurement, is also reported in Table 2. The standard deviation of data describes the dispersion of the data away from the mean, in contrast, the coefficient of variation is the multiple of the standard deviation to the mean, i.e.,   CV . For comparison between data sets with different units or widely different means, we may use the coefficient of variation instead of the standard deviation. And, as described in Scheel (1978) the coefficient of variation can also be a measure of relative risk in the elementary risk and insurance. Such that, an asset with lower value of coefficient of variation means either a lower-risk asset among that of the same return or a higher-return asset among that of same level of risk. As shown in the Table 2, China yuan (CNY) and Singapore dollar (SGD) has lower coefficient of variation, 6.8804 and 8.3299, respectively, and Great British pound and Korean won has higher coefficient of variation. It means that both Great British pound and Korean won are either high-risk or low-return. 3.1. Empirical Results and Analysis First, we have to test whether the series of performance is stationary or not. That is, we should test the null hypothesis that it has a unit root. In Tsay (2005) he indicated that the fundamental time series analysis is stationarity. A time series ty is said to be strictly stationary if the joint distribution of   kttt yyy , , ,  21 is identical to that of   ststst k yyy , , ,  21 for all k, where s is an arbitrary positive integer. In other words, strict stationarity requires that the joint distribution of   kttt yyy , , ,  21 is invariant under time shift. http://www.cbc.gov.tw/content.asp?mp=1&CuItem=36599 Asian Journal of Economics and Empirical Research, 2017, 4(1): 32-48 39 Table-3. The stationarity test results of company‟s performances. Series Obs. ADF test statistic p-value Stationarity 2702 HG ROA 56 -1.539 0.5140 Non-stationary tROA 61 -14.607 0.0000 Stationary 2704 AMBH ROA 56 -2.884 0.0472 Stationary 2705 Leofoo ROA 56 -2.158 0.2217 Non-stationary tROA 61 -12.592 0.0000 Stationary 2706 First Hotel ROA 56 -2.312 0.1683 Non-stationary tROA 61 -17.520 0.0000 Stationary 2707 GFRT ROA 56 -2.640 0.0849 Non-stationary tROA 61 -12.541 0.0000 Stationary 2712 FGH tROA 9 -3.466 0.0089 Stationary 2718 PH ROA 18 -1.651 0.4567 Non-stationary tROA 23 -9.001 0.0000 Stationary 2722 Chateau ROA 14 -1.810 0.3755 Non-stationary tROA 19 -5.904 0.0000 Stationary 2724 FX Hotels ROA 14 0.025 0.9606 Non-stationary tROA 19 -5.816 0.0000 Stationary 5701 JFS ROA 56 -1.476 0.5452 Non-stationary tROA 61 -11.789 0.0000 Stationary 5703 Landis Taipei ROA 56 -1.977 0.2967 Non-stationary tROA 61 -10.758 0.0000 Stationary 5704 Chihpen Royal ROA 56 -1.421 0.5722 Non-stationary tROA 61 -17.149 0.0000 Stationary Source: Taiwan Economic Journal (TEJ). Table-3(B). The stationarity test results of company‟s ROE. Series Obs. ADF test statistic p-value Stationarity 2702 HG ROE 56 -1.604 0.4814 Non-stationary tROE 61 -15.323 0.0000 Stationary 2704 AMBH ROE 56 -2.993 0.0356 Stationary 2705 Leofoo ROE 25 -2.061 0.2604 Non-stationary tROE 30 -8.296 0.0000 Stationary 2706 First Hotel ROE 25 -2.441 0.1306 Non-stationary tROE 30 -13.890 0.0000 Stationary 2707 GFRT ROE 56 -1.808 0.3764 Non-stationary tROE 61 -11.738 0.0000 Stationary 2712 FGH tROE 9 -3.501 0.0080 Stationary 2718 PH ROE 18 -1.689 0.4365 Non-stationary tROE 23 -8.826 0.0000 Stationary 2722 Chateau ROE 14 -2.132 0.2320 Non-stationary tROE 19 -6.034 0.0000 Stationary 2724 FX Hotels ROE 14 0.394 0.9813 Non-stationary tROE 19 -7.304 0.0000 Stationary 5701 JFS ROE 25 -2.505 0.1143 Non-stationary tROE 30 -8.123 0.0000 Stationary 5703 Landis Taipei ROE 56 -1.942 0.3124 Non-stationary tROE 61 -10.652 0.0000 Stationary 5704 Chihpen Royal ROE 56 -1.332 0.6146 Non-stationary tROE 61 -17.278 0.000 Stationary Source: Taiwan Economic Journal (TEJ). Asian Journal of Economics and Empirical Research, 2017, 4(1): 32-48 40 And a time series ty is weakly stationary if both the mean of ty and  stt yyCov , are time-invariant, where s is an arbitrary integer. In the Table 3, we show the Augmented Dicky-Fuller test results. As shown in Table 3, we can find that almost all the ROA/ROE series are non-stationary except the Ambassador‟s ROA/ROE. On the other hand, according to Hurvich and Tsai (1989) there will be biased estimates resulting from a non- stationary series. Such that, applying Wei (2006) we take the first-ordered difference on the series, i.e., 11  ttt ROAROAROAD and 11  ttt ROEROEROED . (18) And then, we test the unit-root-test again to verify its stationarity. The Augmented Dicky-Fuller test results are also shown in Table 3. After differencing the series, all of them are stationary. Next, Patro et al. (2002) found the significant currency risk exposures in country equity index returns by using the GARCH model. And, Polodoo et al. (2016) discussed the nexus between exchange rate volatility and manufacturing trade. They found that exchange rate volatility has an adverse effect on the real manufacturing trade of the Africa countries. As shown in the study of Ikechukwu (2016), he applied the dynamic panel regression approach to investigate the effects of exchange rate volatility on firm performance by examining 20 companies listing in Nigerian Stock Exchange. It revealed that exchange rate volatility has significant negative impacts on the ROAs, ATRs. Here, that effects of the fluctuations of exchange rates on the firm‟s performance is the main purpose of this study. Therefore, as the work in Kim (2012) the autoregression moving average (ARIMA) model can be specified as follows:    p k ktikitmiiti ePerformancDRMRFePerformancD 1 11 , , , ,       q s stistii n j tjji aSizeFX 01 , , , ,  , (19)     iTqpqpt , , , , , 2max1max  , Ni , , , 21 . Here, tiePerformancD , 1 represents the first-ordered difference of the ith firm‟s performance in the tth quarter, and ktiePerformancD  , 1 is its kth lagged variable. Use the STATA13 to find the regression results and show in the Table 4. Model I regresses ROAD1 on all exchange fluctuations, lagged variables and the control variables. Model II regresses ROAD1 on all variables but selected by eliminating higher p-value explanatory variables. Table-4. Regression on ROA. Company Hotel Holiday Garden (2702) The Leofoo Development Co., Ltd. (2705) Formosa International Hotels Corporation (2707) Variables Model I Model II Model I Model II Model I Model II Const. 10.31 (7.59) 0.10 (0.09) 106.98 (81.09) 101.70** (47.14) 23.71 (16.21) -0.06 (0.14) RMRF 0.01 (0.01) -0.05 (0.07) -0.01 (0.02) USD 0.14 (0.14) 0.15 (0.33) 0.10 (0.20) JPY 0.03 (0.03) -0.01 (0.09) -0.05 (0.05) CNY -0.09 (0.14) -0.38 (0.36) -0.21 (0.21) EUR -0.05 (0.05) 0.09 (0.13) -0.00 (0.08) KRW 0.02 (0.04) 0.28* (0.14) 0.20** (0.08) 0.04 (0.07) GBP 0.03 (0.05) -0.12 (0.17) -0.03 (0.08) SGD 0.05 (0.14) 0.51 (0.62) 0.41* (0.21) 0.06 (0.21) AUD -0.04 (0.04) -0.04* (0.02) -0.21 (0.19) -0.20** (0.07) 0.06 (0.06) 0.09*** (0.03) IDR 0.03 (0.03) 0.05** (0.02) 0.01 (0.13) -0.09* (0.05) -0.07** (0.03) THB 0.05 (0.06) 0.22 (0.20) -0.00 (010) MYR -0.03 (0.07) -0.17 (0.21) -0.01 (0.11) PHP -0.04 (0.06) -0.07 (0.18) 0.09 (0.09) Lag1 -0.83*** (0.18) 0.67*** (0.13) -0.91** (0.33) -0.76*** (0.18) -0.69*** (0.15) -0.57*** (0.12) Lag2 -0.63*** (0.20) -0.49*** (0.14) -0.34 (0.34) -0.35** (0.17) -0.80*** (0.16) -0.67*** (0.12) Lag 3 -0.45** (0.19) -0.45*** (0.11) 0.01 (0.33) -0.51*** (0.15) -0.35*** (0.11) Lag4 0.01 (0.15) -0.04 (0.25) -0.13 (0.15) SIZE -0.53 (0.36) -4.70 (3.57) -4.48** (2.08) -1.06 (0.73) Adj. R2 0.46 0.51 0.03 0.32 0.38 0.44 Obs. 59 58 32 32 58 58 Source: Taiwan Economic Journal (TEJ). Asian Journal of Economics and Empirical Research, 2017, 4(1): 32-48 41 The regression model is given above. . , , , , , , , , titii n j tjji p k ktikitmiiti aSizeFXROADRMRFROAD      11 11 (20) Model I regresses 1,,,1   ttt ROAROAROAD on all exchange fluctuations, lagged variables and the control variables. And Model II regresses ROAD1 on all variables but selected by eliminating higher p-value explanatory variables. The values in the parentheses are standard error of the estimates. And *, ** and *** stand for 10%, 5% and 1% level of significance, respectively. Table-4. Regression on ROA (Continued). Company First Hotel Company Ltd. (2706) Pleasant Hotels International Inc. (2718) Chateau International Development Co., Ltd. (2722) Variables Model I Model II Model I Model II Model I Model II Const. -8.17 (15.86) -14.17* 37.34 (58.01) 0.71*** (0.12) -1957** (542.5) 0.29 (0.85) RMRF 0.01 (0.01) -0.12 (0.10) -0.06** (0.03) 3.33*** (0.79) USD 0.06 (0.05) 1.54 (1.07) 0.66*** (0.15) -13.15** (4.20) JPY 0.01 (0.01) -0.30 (0.12) 3.14*** (0.76) CNY -0.05 (0.06) -1.36 (0.72) -0.78*** (0.14) 12.03** (3.78) EUR 0.00 (0.02) -0.29 (0.20) -0.16*** (0.04) 7.60** (1.96) KRW -0.00 (0.02) 0.23 (0.23) -5.31** (1.49) GBP -0.02 (0.03) -0.26 (0.38) -6.44*** (1.50) SGD -0.05 (0.07) -0.66 (0.43) -0.66** (0.11) 1.41 (0.74) 1.89** (0.85) AUD -0.01 (0.02) 0.21 (0.13) 0.13** (0.05) -4.36*** (0.91) -0.58* (0.29) IDR 0.03 (0.02) 0.20 (0.08) 0.22*** (0.03) 0.59* (0.24) THB 0.00 (0.04) 0.58 (0.32) 0.55*** (0.12) -13.42*** (3.22) MYR -0.02 (0.03) -0.06 (0.16) 3.75** (0.96) PHP -0.04 (0.03) -0.04*** (0.01) -1.06 (0.43) 18.45** (4.82) Lag1 -1.05*** (0.22) -1.21*** (0.14) -1.90 (0.46) -1.46*** (0.10) Lag2 -0.28 (0.24) -1.46 (0.44) -1.13*** (0.12) Lag 3 -0.47 (0.27) -0.98 (0.63) -0.44*** (0.10) Lag4 0.34*** (0.10) -0.30*** (0.07) -0.22 (0.25) SIZE 0.41 (0.72) 0.67* (0.34) -1.80 (2.85) 92.79** (25.69) Adj. R2 0.81 0.82 0.84 0.93 0.78 0.16 Obs. 32 32 20 20 20 20 Source: Taiwan Economic Journal (TEJ). In the Table 4, we can find that almost all estimates of the lagged variables are significant and negative, such as, Leofoo Development Co., Ltd. (2705), Formosa International Hotels Corporation (2707), Janfusun Fancyworld Corp. (5701), The Landis Taipei Hotel Co., Ltd. (5703), and Hotel Royal Chihpen (5704). It implies that those ROAD1 are mean-reverting. As the estimates of third-lagged variables are also significant, then we can conclude that there is a seasonal effect on the company‟s ROA. Moreover, some estimates of SIZE are significant in Table 4. When it is positive, such as that in Chateau International Development Co., Ltd. (2722), the company may increase its own assets to increase its ROAD1 , so to its ROA, too. Since it can operate efficiently its assets to generate more profit and then to be a well-performed company. On the other hand, when the estimate of SIZE is negative, such as those in Leofoo Development Co., Ltd. (2705) and FX Hotels Group Inc. (2724-F), the company may dispose some of its idle assets or non-performed assets to reduce the inefficient effect of these assets. As a result, the company‟s ROA will be improved. Asian Journal of Economics and Empirical Research, 2017, 4(1): 32-48 42 Table-4. Regression on ROA (Continued). Company Janfusun Fancyworld Corp. (5701) The Landis Taipei Hotel Co., Ltd. (5703) Hotel Royal Chihpen (5704) Variables Model I Model II Model I Model II Model I Model II Const. 29.91 (30.30) -0.12 (0.18) 55.80 (50.46) 0.13 (0.17) 32.16* (19.05) 0.00 (0.12) RMRF 0.04 (0.05) 0.06 (0.03) 0.07*** (0.02) 0.05** (0.02) 0.03** (0.01) USD -0.24 (0.31) 0.10 (0.24) 0.07 (0.17) JPY -0.01 (0.08) -0.11 (0.08) 0.07 (0.05) CNY 0.38 (0.36) 0.10 (0.27) 0.22 (0.19) 0.22*** (0.07) EUR 0.03 (0.12) -0.11 (0.10) -0.12* (0.07) 0.01 (0.07) KRW 0.08 (0.12) 0.10* (0.05) 0.07 (0.09) 0.27*** (0.07) 0.17*** (0.04) GBP -0.24 (0.15) -0.16** (0.06) 0.00 (0.10) -0.09 (0.07) SGD 0.29 (0.37) -0.05 (0.24) -0.31* (0.18) -0.29*** (0.09) AUD -0.10 (0.12) 0.22** (0.09) 0.20*** (0.06) 0.02 (0.06) IDR 0.08 (0.11) -0.02 (0.06) -0.01 (0.04) THB -0.21 (0.18) -0.12 (0.12) -0.20** (0.08) 0.00 (0.09) MYR 0.06 (0.17) -0.17 (0.13) 0.04 (0.10) PHP 0.06 (0.15) -0.02 (0.10) -0.10 (0.07) Lag1 -0.65** (0.28) -0.64*** (0.16) -0.77*** (0.12) -0.77*** (0.09) -1.08*** (0.15) -0.99*** (0.09) Lag2 -0.31 (0.36) -0.31* (0.16) -0.32** (0.14) -0.35*** (0.09) -0.83*** (0.18) -0.66*** (0.12) Lag 3 0.09 (0.37) 0.01 (0.13) -0.71*** (0.19) -0.56*** (0.09) Lag4 0.14 (0.26) -0.07 (0.11) -0.11 (0.16) SIZE -1.35 (1.36) -2.68 (2.41) -1.58* (0.93) Adj. R2 0.07 0.39 0.59 0.63 0.75 0.76 Obs. 32 32 58 58 58 59 Source: Taiwan Economic Journal (TEJ). Next, Table 4 shows significant effects on the performances of Taiwan tourism industry due to the fluctuations of foreign exchange rates. The changes of foreign exchange rates have significant impacts on the D1ROAs. Some are positive and some are negative. And the same currency has different impact on different companies. Such as the Singapore dollar has positive effect on the D1ROA of Leofoo Development Co., Ltd. (2705), Chateau International Development Co., Ltd. (2722), and on the ROA of Ambassador Hotel Ltd. (2704), but negative effect on that of Pleasant Hotels International Inc. (2718) and Hotel Royal Chihpen (5704). Moreover, the Australian dollar has positive effect on the D1ROA of Formosa International Hotels Corporation (2707), Pleasant Hotels International Inc. (2718), and Landis Taipei Hotel Co., Ltd. (5703), and on the ROA of Ambassador Hotel Ltd. (2704), but negative effect on that of Chateau International Development Co., Ltd. (2722). And the Korean won has a positive effect on the D1ROA of Leofoo Development Co., Ltd. (2705), Janfusun Fancyworld Corp. (5701), and Hotel Royal Chihpen (5704), and then on those company‟s ROA , too. Furthermore, the number of significant variables and the component of significant variables are different to each company. For example, the significant variables of the Pleasant‟s ROAD1 are the change of USD, CNY, EUR, SGD, AUD, IDR, THB, however, that of the Chateau‟s D1ROA are only the changes of Singapore dollar and Australia dollar. As a result, the portfolio of currencies should be different for each company. Asian Journal of Economics and Empirical Research, 2017, 4(1): 32-48 43 Table-5. Regression on ROE. The regression model is given as follows: . , , , , , , , , titii n j tjji p k ktikitmiiti aSizeFXROEDRMRFROED      11 11 (21) Model I regresses 1,,,1   ttt ROEROEROED on all exchange fluctuations, lagged variables and the control variables. And Model II regresses ROED1 on all variables but selected by eliminating higher p-value explanatory variables. The values in the parentheses are standard error of the estimates. And *, ** and *** stand for 10%, 5% and 1% level of significance, respectively. Company Hotel Holiday Garden (2702) The Leofoo Development Co., Ltd. (2705) Formosa International Hotels Corporation (2707) Variables Model I Model II Model I Model II Model I Model II Const. 8.39 (10.97) 0.14 (0.13) 550.62 (305.52) 232.88** (110.24) 28.18 (15.24) 0.10 (0.21) RMRF 0.01 (0.02) -0.07 (0.21) -0.03 (0.04) USD 0.24 (0.19) -1.77 (1.44) 0.05 (0.31) JPY 0.05 (0.05) 0.08** (0.03) -0.64* (0.33) -0.39** (0.14) -0.12 (0.08) -0.09* (0.05) CNY -0.16 (0.20) 1.37 (1.40) -0.27 (0.33) -0.25** (0.11) EUR -0.11 (0.08) -0.07* (0.04) -0.00 (0.34) -0.10 (0.13) KRW 0.05 (0.06) 0.08* (0.04) 0.47 (0.46) 0.08 (0.11) GBP 0.00 (0.07) -0.19 (0.42) -0.04 (0.12) SGD 0.22 (0.20) 4.02** (1.70) 1.74** (0.62) 0.34 (0.34) 0.39** (0.19) AUD -0.00 (0.20) -0.63 (0.50) 0.06 (0.10) IDR 0.02 (0.05) 0.12 (0.28) -0.13* (0.07) -0.09* (0.05) THB 0.11 (0.09) 0.10* (0.06) -0.52 (0.77) 0.02 (0.16) MYR -0.06 (0.11) -1.36** (0.56) -0.68** (0.29) -0.01 (0.18) PHP -0.11 (0.08) 0.92 (1.24) 0.05 (0.13) Lag1 -0.90*** (0.15) -0.88*** (0.12) -1.27*** (0.37) -0.64*** (0.18) -0.61*** (0.15) -0.61*** (0.12) Lag2 -0.67*** (0.19) -0.58** (0.15) -0.15 (0.43) -0.66*** (0.16) -0.63*** (0.12) Lag 3 0.41** (0.19) -0.42*** (0.18) 0.20 (0.33) -0.49*** (0.16) -0.46*** (0.12) Lag4 -0.04 (0.15) 0.33 (0.28) -0.01 (0.16) SIZE -0.39 (0.57) -24.32 (13.46) -10.28** (4.86) -1.26 (1.13) Adj. R2 0.51 0.54 0.16 0.32 0.30 0.39 Obs. 58 58 27 27 58 58 Source: Taiwan Economic Journal (TEJ). Asian Journal of Economics and Empirical Research, 2017, 4(1): 32-48 44 Table-5. Regression on ROE (Continued). Company First Hotel Company Ltd. (2706) Pleasant Hotels International Inc. (2718) Chateau International Development Co., Ltd. (2722) Variables Model I Model II Model I Model II Model I Model II Const. -21.59 (37.72) 0.84** (0.38) 31.69 (89.17) 0.98*** (0.18) -2289** (737.7) -0.44 (1.05) RMRF 0.05* (0.02) 0.05*** (0.01) -0.15 (0.15) -0.08* (0.04) 3.87** (1.08) USD 0.39** (1.74) 0.12*** (0.04) 1.77 (1.66) 0.85*** (0.21) -15.42** (5.71) JPY 0.04 (0.03) -0.02 (0.18) 3.60** (1.03) CNY -0.31* (0.16) -1.61 (1.12) -1.04*** (0.21) 14.08** (5.14) EUR -0.05 (0.05) -0.37 (0.31) -0.24*** (0.06) 8.86** (2.67) KRW -0.04 (0.05) 0.24 (0.35) -6.07** (2.03) GBP 0.01 (0.06) -0.35 (0.63) -7.56** (2.03) SGD 0.03 (0.14) -0.76 (0.72) -0.86*** (0.15) 1.93 (1.00) 2.46** (1.05) AUD 0.03 (0.05) 0.25 (0.21) 0.16** (0.07) -5.22*** (1.24) -0.77** (0.36) IDR -0.04 (0.05) 0.27 (0.13) 0.29*** (0.05) 0.77* (0.32) THB 0.05 (0.07) 0.77 (0.50) 0.74*** (0.17) -15.90** (4.38) MYR -0.04 (0.06) -0.13 (0.25) 4.41** (1.30) PHP -0.08 (0.11) -1.32 (0.67) -0.93*** (0.23) 21.70** (6.56) Lag1 -1.37*** (0.18) -2.88*** (0.37) -1.88 (0.55) -1.48*** (0.12) Lag2 -0.31 (0.21) -0.82** (0.40) -1.50 (0.55) -1.15*** (0.13) Lag 3 -0.19 (0.24) -0.99 (0.80) -0.44*** (0.11) Lag4 0.50** (0.21) 1.86*** (0.36) -0.21 (0.29) SIZE 1.06 (1.74) -1.51 (4.37) 108.51** (34.93) Adj. R2 0.87 0.85 0.77 0.92 0.74 0.18 Obs. 28 31 20 20 20 20 Source: Taiwan Economic Journal (TEJ). In Table 5, Model I regresses D1ROE on all exchange fluctuations, lagged variables and the control variables. Model II regresses D1ROE on all variables but selected by eliminating higher p-value explanatory variables. We may find that the results in Table 5 are almost the same as in Table 4. There is seasonal effect for Taiwan hotel industry‟s ROE, too. And the D1ROE of First Hotel Company Ltd. (2706) and Pleasant Hotels International Inc. (2718) are mean-reverting. Moreover, the number of significant variables and the component of significant variables are different to each company. For example, the significant variables of the ROE of Landis Taipei Hotel Co., Ltd. (5703) are the changes of euro, Japan yen, Australia dollar and Malaysian Ringgit, but that of the Chateau International Development Co., Ltd. (2722) are the changes of euro, pound, Chinese yuan, Japan yen, Korean won, Singapore dollar, Australia dollar, Thailand Baht, Malaysian Ringgit, and Philippine peso. Therefore, it supports the results in the Table 4, which the portfolio of currencies should be different for each company. Asian Journal of Economics and Empirical Research, 2017, 4(1): 32-48 45 Table-5. Regression on ROE (Continued). Company Janfusun Fancyworld Corp. (5701) The Landis Taipei Hotel Co., Ltd. (5703) Hotel Royal Chihpen (5704) Variables Model I Model II Model I Model II Model I Model II Const. 55.72 (141.90) 0.04 (0.42) 70.20 (69.75) -0.28 (0.23) 39.26* (21.32) 0.02 (0.14) RMRF 0.09 (0.33) 0.07 (0.04) 0.06** (0.02) 0.04** (0.02) USD 0.11 (1.45) 0.17 (0.33) 0.04 (0.19) JPY 0.12 (0.35) -0.15 (0.10) -0.15** (0.06) 0.08 (0.05) CNY 0.01 (1.23) 0.11 (0.37) 0.28 (0.21) 0.25*** (0.08) EUR 0.12 (0.45) -0.15 (0.14) -0.25*** (0.09) 0.02 (0.08) KRW 0.37 (0.72) 0.08 (0.12) 0.30*** (0.07) 0.18*** (0.05) GBP -0.59 (0.43) -0.31** (0.14) -0.02 (0.14) -0.09 (0.08) SGD 0.16 (1.00) -0.09 (0.33) -0.36* (0.20) -0.32*** (0.10) AUD -0.51 (0.37) -0.22* (0.11) 0.30** (0.12) 0.39*** (0.07) 0.01 (0.06) IDR 0.23 (0.33) -0.03 (0.08) -0.01 (0.05) THB -0.42 (0.66) -0.11 (0.17) -0.00 (0.10) MYR 0.51 (0.48) 0.36* (0.20) -0.23 (0.18) -0.24** (0.11) 0.05 (0.11) PHP 0.20 (1.13) -0.03 (0.14) -0.11 (0.08) Lag1 -0.60* (0.31) -0.75*** (0.17) -0.76*** (0.13) -0.75*** (0.09) -1.09*** (0.14) -0.99*** (0.09) Lag2 -0.29 (0.43) -0.52*** (0.17) -0.35** (0.15) -0.42*** (0.09) -0.83*** (0.19) -0.66*** (0.12) Lag 3 0.30 (0.73) -0.01 (0.13) -0.73*** (0.19) -0.56*** (0.09) Lag4 0.26 (0.48) -0.08 (0.11) -0.12 (0.16) SIZE -2.50 (6.39) -3.37 (3.33) -1.93* (1.05) Adj. R2 0.04 0.48 0.57 0.61 0.76 0.77 Obs. 27 27 58 58 58 58 Source: Taiwan Economic Journal (TEJ). Next, we‟ll analyze the portfolio will be foreign exchange position. Because of our foreign trade is mainly denominated in US dollars, but also long-term focus Taipei currency exchange on the USD/NTD, and therefore may have a greater proportion of dollar holdings. Furthermore, since the Chinese mainland tourists to Taiwan surge trips, so that each of the hotel were increased demand for Chinese yuan transactions, and thus the performance of the reaction in the performance of its ROE or ROA. On the other hand, Taiwan is also the first choice for Japanese and Korean tourists traveling abroad, so accommodation for the Korean won and the Japanese yen in trading volume should not be underestimated. As shown in Table 7, the 2012 tourist‟s sources distribution for Taiwan major hotels aggregated by the Tourism Bureau, MOTC of Taiwan, the Japanese and Korean inbounds are over 1/5 of guests in the half of the hotels. And as the Pleasant Hotel locates at Taoyuan, closed to the Taoyuan International Airport, such that, most Chinese mainland tourists stay at the hotel in order to entry and exit. Jang and Chen (2008); Chen et al. (2011) employed the modern portfolio theory to investigate the mixes of inbounds of Taiwan inbounds. They suggested that the government should take the high-reward/high-volatility option and shift more available resources to attract the Japanese tourists. Asian Journal of Economics and Empirical Research, 2017, 4(1): 32-48 46 Table-6. Regression on Financial Performances of the Ambassador Hotel. The regression model is given as follows: . , , , , , , , , tti n j tjj p k ktktmt aSizeFXyRMRFy 27042704 1 2704 1 27042704270427042704      (22) The dependent variable, y2704 represents the performance of the Ambassador Hotel, which is either ROA2704 or ROE2704. Model I regresses y2704 on all exchange fluctuations, lagged variables and the control variables. And Model II regresses y2704 on all variables but selected by eliminating higher p-value explanatory variables. The values in the parentheses are standard error of the estimates. And *, ** and *** stand for 10%, 5% and 1% level of significance, respectively. Performance ROA2704 ROE2704 Variables Model I Model II Model I Model II Const. -25.80 (34.26) 0.28** (0.11) -34.80 (58.29) 0.25* (0.14) RMRF 0.02 (0.01) 0.15** (0.01) 0.03 (0.02) 0.03*** (0.01) USD -0.21** (0.09) -0.32** (0.15) JPY 0.01 (0.02) 0.02 (0.03) CNY 0.22** (0.09) 0.34** (0.16) EUR -0.05 (0.03) -0.05** (0.02) -0.08 (0.06) -0.08** (0.03) KRW 0.01 (0.03) 0.03 (0.05) GBP 0.04 (0.03) 0.06 (0.06) SGD 0.10 (0.10) 0.14** (0.06) 0.17 (0.17) 0.26** (0.10) AUD -0.02 (0.03) -0.05 (0.05) IDR 0.01 (0.02) 0.02 (0.04) THB 0.02 (0.04) 0.04 (0.07) MYR -0.11** (0.05) -0.10*** (0.04) -0.18** (0.08) -0.17*** (0.06) PHP 0.02 (0.04) 0.01 (0.07) Lag1 0.20 (0.14) 0.26** (0.12) 0.30** (0.14) 0.32*** (0.11) Lag2 -0.06 (0.16) -0.06 (0.15) Lag 3 -0.12 (0.14) -0.10 (0.14) Lag4 0.25* (0.14) 0.28** (0.11) 0.24* (0.14) 0.24** (0.11) SIZE 1.13 (1.48) 1.52 (2.52) Adj. R2 0.27 0.40 0.30 0.36 Obs. 59 59 59 59 Source: Taiwan Economic Journal (TEJ). Table-7. Distribution of guests‟ sources in 2012. Hotel Region Royal Hotel Pleasant Hotels (Taoyuan) Ambassador Hotel Landis Taipei Hotel Formosa International Hotels Leofoo Westin Hotel Holiday Garden Hotel Farglory Hotel Domestic 55.16 17.99 33.96 24.15 21.33 9.73 61.24 94.92 Oversea Chinese 0.00 7.65 1.91 6.37 0.00 0.00 0.99 0.00 Mainland 6.35 56.35 13.29 11.08 11.52 19.43 22.45 3.86 North American 4.73 0.29 6.94 10.31 8.42 20.81 0.76 0.20 Japan 21.93 2.09 29.85 29.89 36.67 17.27 7.00 0.10 Asian (exclusive Japanese) 5.00 9.21 8.23 8.39 15.25 25.92 6.54 0.48 European 2.51 0.26 3.77 7.53 4.69 4.33 0.51 0.13 Australia 0.28 0.03 0.41 1.83 0.90 1.35 0.53 0.02 Others 4.05 6.13 1.64 0.44 1.23 1.17 0.00 0.30 Total (%) 100 100 100 100 100 100 100 100 Source: Tourism Bureau, M.O.T.C., Republic of China (Taiwan). Here, refer to Kim (2013) discussion of foreign exchange position to make recommendations in the following table. According to the analysis results in Table 4, 5, and 6, we can form a portfolio of currencies that have Asian Journal of Economics and Empirical Research, 2017, 4(1): 32-48 47 significant impacts on the company‟s ROA/ ROE. Using the Modern portfolio theory proposed by Markowitz (1952) based on the weighted each company the average cost of capital (abbreviated as WACC), and calculated by Matlab programs for foreign exchange positions, we may find an optimum allocation of currencies which has the lowest degree of risk under a pre-specified rate of return constraint. Table-8. Optimal Portfolio of Foreign Currencies for each company. SEC_id USD JPY GBP EUR KRW SGD CNY AUD IDR THB MYR PHP Required Return (%) Portfolio Risk 2702 2.32 20.47 3.84 52.35 1.73 19.28 10.0 0.4305 2704 13.79 84.35 1.85 12.0 1.8498 2705 1.30 1.56 3.84 83.81 9.49 13.5 0.5125 2706 0.00 2.09 62.18 30.56 0.00 5.17 0.00 12.8 2.5007 2707 0.00 0.00 19.47 74.37 6.16 9.5 0.3555 2718 6.35 0.00 0.00 73.79 7.14 0.00 5.53 7.19 12.6 0.4580 2722 3.27 0.00 6.07 0.00 5.17 85.48 10.6 0.5154 2724 2.26 0.00 0.31 0.00 4.47 82.84 5.47 4.66 9.8 0.4807 5701 2.24 0.76 85.53 11.47 13.2 0.5204 5703 0.11 6.04 79.93 0.90 13.03 11.5 0.4757 5704 34.33 14.90 0.00 13.80 36.97 9.8 2.0615 Full 0.00 0.00 0.29 44.14 0.07 0.00 1.49 9.41 0.52 0.00 0.64 43.44 12.2 0.0554 Source: Taiwan Economic Journal (TEJ). The results in Table 8 show the optimal allocation of currencies for each company. Here, we can find that Japanese yen, Korean won, Chinese Yuan, Australian dollar and Malaysian Ringgit configuration still the majority. Among them, Leofoo Development Co., Ltd. (2705), Formosa International Hotels Corporation (2707), Pleasant Hotels International Inc. (2718), Chateau International Development Co., Ltd. (2722), FX Hotels Group Inc. (2724-F), Janfusun Fancyworld Corp. (5701), The Landis Taipei Hotel Co., Ltd. (5703), in the configuration of the Australian dollar reached 52.35%, 83.81%, 44.37%, 73.79%, 85.48%, 82.84%, 85.53% and 79.93 %, respectively, more than 50% have switched. The Hotel Holiday Garden (2702), The Leofoo Development Co., Ltd. (2705), First Hotel Company Ltd. (2706), Chateau International Development Co., Ltd. (2722), FX Hotels Group Inc. (2724-F), Janfusun Fancyworld Corp. (5701), and Landis Taipei Hotel Co., Ltd. (5703) for the Korean won configuration, respectively, 3.84%, 1.56%, 30.56%, 6.07%, 0.31%, 0.76%, and 14.90. As to the Chinese yuan, Formosa International Hotels Corporation (2707), Chateau International Development Co., Ltd. (2722), FX Hotels Group Inc. (2724-F), and Hotel Royal Chihpen (5704) should put the weight ranging from 4.47% to 19.47%. 4. Conclusions Recent years, changes in exchange rates will significantly affect the performances of a company, such as, ROE, ROA, etc. Faced with dramatic changes in the international economic environment, as well as central banks continue to adopt a more aggressive monetary policy, such as: Bank of Japan negative interest rates, the ECB‟s monetary easing, China People‟s Bank of China monetary easing, and the gradual recovery of the economy of the United States have taken action to raise interest rates and so on. Under the auspices of monetary policy in these countries, it shows once again that the currencies flows across countries and international hot money have allowed changes in exchange rates and more intense. And Taiwanese enterprises face to these monetary policies, foreign exchange positions should be actively managed in order to reduce the impacts suffered. In the past, the fluctuations in the foreign exchange markets are more stable in today. In addition to monetary policies that attract more investors to the market, the investment of foreign exchange market as well significantly affect the change in exchange rates among countries. Therefore, a positive foreign exchange risk management will better help for future operation, which can significantly reduce the risk of foreign exchange movements. This study found that Taiwanese hospitality companies, accounting for the largest part of the tourism industry, are subject to have the impacts on their performance and profitability due to the exchange rate fluctuations. Multinational enterprises may apply the results developed here to manage their foreign exchange risk exposure, and then increasing the overall capacity and range of enterprise risk management (ERM). By doing so, corporate can increase their profits and reduce the negative impacts of exchange rate changes on corporate ROE/ROA through foreign exchange operations. More importantly, foreign exchange allocation can be a strategy to reduce the risk of foreign exchange exposure. References Adler, M. and D. Bernard, 1980. The exposure of long-term foreign currency bonds. Journal of Financial and Quantitative Analysis, 15(4): 973-994. View at Google Scholar | View at Publisher Bailey, W., E. Ng and R.M. Stulz, 1992. Optimal hedging of stock portfolios against foreign exchange risk: Theory and applications. Global Finance Journal, 3(2): 97-113. View at Google Scholar | View at Publisher Bodie, Z., A. Kane and A. Marcus, 2002. Investment. New York: McGraw Hill. Bodnar, G.M. and G.M. William, 1993. Exchange rate exposure and industry characteristics: Evidence from Canada, Japan, and the USA. Journal of International Money and Finance, 12(1): 29-45. View at Google Scholar | View at Publisher Chen, C.-F. and C.-W.S. Zan, 2009. Tourism expansion, tourism uncertainty and economic growth: New evidence from Taiwan and Korea. Tourism Management, 30(6): 812–818. View at Google Scholar | View at Publisher Chen, M.H., J. SooCheong and Y.J. Peng, 2011. Discovering optimal tourist market mixes. Journal of Travel Research, 50(6): 602-614. View at Google Scholar | View at Publisher Dumas, B., 1978. The theory of the trading firm revisited. Journal of Finance, 33(3): 1019-1030. View at Google Scholar | View at Publisher Elton, E., M. Gruber, S. Brown and W. Goetzmann, 2007. Modern portfolio theory and investment analysis. New York: Wiley. Fama, E.F. and F.R. Kenneth, 1993. Common risk factors in the returns on stocks and bonds. Journal of Financial Economics, 33(1): 3-56. View at Google Scholar | View at Publisher Fama, E.F. and F.R. Kenneth, 1995. Size and book-to-market factors in earnings and returns. Journal of Finance, 50(1): 131-155. View at Google Scholar | View at Publisher Hamid, A.H., K.H. Siarap and A. Mustafa, 2013. Evaluation of a public relations agency (PRA)‟s performance of promoting tourism using importance performance analysis (IPA). Journal of Tourism and Hospitality Management, 1(2): 95-102. View at Publisher https://scholar.google.com/scholar?hl=en&q=The%20exposure%20of%20long-term%20foreign%20currency%20bonds http://dx.doi.org/0.2307/2330573 https://scholar.google.com/scholar?hl=en&q=Optimal%20hedging%20of%20stock%20portfolios%20against%20foreign%20exchange%20risk:%20Theory%20and%20applications http://dx.doi.org/10.1016/1044-0283(92)90007-8 https://scholar.google.com/scholar?hl=en&q=Exchange%20rate%20exposure%20and%20industry%20characteristics:%20Evidence%20from%20Canada,%20Japan,%20and%20the%20USA http://dx.doi.org/10.1016/0261-5606(93)90008-y https://scholar.google.com/scholar?hl=en&q=Tourism%20expansion,%20tourism%20uncertainty%20and%20economic%20growth:%20New%20evidence%20from%20Taiwan%20and%20Korea http://dx.doi.org/10.1016/j.tourman.2008.12.013 https://scholar.google.com/scholar?hl=en&q=Discovering%20optimal%20tourist%20market%20mixes https://scholar.google.com/scholar?hl=en&q=Discovering%20optimal%20tourist%20market%20mixes http://dx.doi.org/10.1177/0047287510385464 https://scholar.google.com/scholar?hl=en&q=The%20theory%20of%20the%20trading%20firm%20revisited http://dx.doi.org/10.1111/j.1540-6261.1978.tb02041.x https://scholar.google.com/scholar?hl=en&q=Common%20risk%20factors%20in%20the%20returns%20on%20stocks%20and%20bonds http://dx.doi.org/10.1016/0304-405x(93)90023-5 https://scholar.google.com/scholar?hl=en&q=Size%20and%20book-to-market%20factors%20in%20earnings%20and%20returns https://scholar.google.com/scholar?hl=en&q=Size%20and%20book-to-market%20factors%20in%20earnings%20and%20returns http://dx.doi.org/10.1111/j.1540-6261.1995.tb05169.x http://dx.doi.org/10.17265/2328-2169/2013.11.005 Asian Journal of Economics and Empirical Research, 2017, 4(1): 32-48 48 He, J. and L.K. Ng, 1998. The foreign exchange exposure of Japanese multinational corporations. Journal of Finance, 53(2): 733-753. View at Google Scholar | View at Publisher Hodder, J., 1982. Exposure to exchange rate movements. Journal of International Economics, 13(3-4): 375-386. View at Google Scholar | View at Publisher Huang, C.F. and R. Litzenberger, 1988. Foundations for financial economics. New-York: North-Holland. Hurvich, C.M. and C.L. Tsai, 1989. Regression and time series model selection in small samples. Biometrika, 76(2): 297-307. View at Google Scholar | View at Publisher Ikechukwu, K., 2016. Exchange rate volatility and firm performance in Nigeria: A dynamic panel regression approach. Journal of Developing Areas, 50(6): 161-174. View at Google Scholar | View at Publisher Jang, S. and M.H. Chen, 2008. Financial portfolio approach to optimal tourist market mixes. Tourism Management, 29(4): 761–770. View at Google Scholar | View at Publisher Jorion, P., 1990. The exchange rate exposure of U.S. multinationals. Journal of Business, 63(3): 331-345. View at Google Scholar | View at Publisher Kim, H.Y., M.H. Chen and S. Jang, 2006. Tourism expansion and economic development: The case of Taiwan. Tourism Management, 27(5): 925–933. View at Publisher Kim, Y.-Y., 2013. Optimal foreign exchange risk hedging: A mean-variance portfolio approach. Theoretical Economics Letters, 3(1): 1-6. View at Google Scholar | View at Publisher Kim, Y.Y., 2012. Stationary vector autoregressive representation of error correction models. Theoretical Economics Letters, 2(2): 152-156. View at Google Scholar | View at Publisher Maloney, P.J., 1990. Management currency exposure: The case of Western mining. Journal of Applied Corporate Finance, 2(4): 29-34. View at Google Scholar | View at Publisher Markowitz, H.M., 1952. Portfolio selection. Journal of Finance, 7(1): 77-91. View at Google Scholar Min, C.K., 2013. Time-specific effects and a test of the tourism-led growth hypothesis. Journal of Tourism and Hospitality Management, 1(3): 133-139. Morelli, D., 2007. Beta, size, book-to-market equity and returns: A study based on UK data. Journal of Multinational Financial Management, 17(3): 257-272. View at Google Scholar | View at Publisher Oh, C.O., 2005. The contribution of tourism development to economic growth in the Korean economy. Tourism Management, 26(1): 39-44. View at Google Scholar | View at Publisher Patro, D.K., J.K. Wald and Y. Wu, 2002. Explaining exchange rate risk in world stock markets: A panel approach. Journal of Banking & Finance, 26(10): 1951-1972. View at Google Scholar | View at Publisher Polodoo, V., B. Seetanah and R.B. Sannassee, 2016. Exchange rate volatility and manufacturing trade: Evidence from Africa. Journal of Developing Areas, 50(6): 133-148. View at Google Scholar | View at Publisher Portnov, B.A. and W. Li, 2013. Investigating the effect of global economic crisis on foreign tourism to Taiwan. Journal of Tourism and Hospitality Management, 1(3): 113-132. Pritamani, M., D. Shome and V. Singal, 2005. Exchange rate exposure of exporting and importing firms. Journal of Applied Corporate Finance, 17(3): 87-94. View at Google Scholar | View at Publisher Scheel, W.C., 1978. Comparisons of riskiness as measured by the coefficient of variation. Journal of Risk and Insurance, 45(1): 148-152. View at Google Scholar | View at Publisher Sharpe, W.F., 1964. Capital asset pricing: A theory of market equilibrium under conditions of risk. Journal of Finance, 19(3): 425-442. View at Google Scholar | View at Publisher Smithson, C. and B.J. Simkins, 2005. Does risk management add value? A survey of the evidence. Journal of Applied Corporate Finance, 17(3): 8-17. View at Google Scholar | View at Publisher Tsay, R.S., 2005. Analysis of financial time series. 2nd Ed., Chapter 2: Liner Time Series Analysis and Its Applications. John Wiley & Sons, Inc. pp: 24-96. Wei, W.W.S., 2006. Time series analysis: Univariate and multivariate methods. 2nd Ed., Chapter 4: Nonstationary Time Series Models, Addison-Wesley. pp: 68-86. Witt, S.F. and C. Witt, 1992. Modeling and forecasting demand in tourism. London: Academic Press. Witt, S.F. and C. Witt, 1995. Forecasting tourism demand: A review of empirical research. International Journal of Forecasting, 11(3): 447- 475. View at Google Scholar | View at Publisher Asian Online Journal Publishing Group is not responsible or answerable for any loss, damage or liability, etc. caused in relation to/arising out of the use of the content. Any queries should be directed to the corresponding author of the article. https://scholar.google.com/scholar?hl=en&q=The%20foreign%20exchange%20exposure%20of%20Japanese%20multinational%20corporations https://scholar.google.com/scholar?hl=en&q=The%20foreign%20exchange%20exposure%20of%20Japanese%20multinational%20corporations http://dx.doi.org/10.1111/0022-1082.295575 https://scholar.google.com/scholar?hl=en&q=Exposure%20to%20exchange%20rate%20movements http://dx.doi.org/10.1016/0022-1996(82)90065-4 http://dx.doi.org/10.1016/0022-1996(82)90065-4 https://scholar.google.com/scholar?hl=en&q=Regression%20and%20time%20series%20model%20selection%20in%20small%20samples https://scholar.google.com/scholar?hl=en&q=Regression%20and%20time%20series%20model%20selection%20in%20small%20samples http://dx.doi.org/10.1093/biomet/76.2.297 https://scholar.google.com/scholar?hl=en&q=Exchange%20rate%20volatility%20and%20firm%20performance%20in%20Nigeria:%20A%20dynamic%20panel%20regression%20approach http://dx.doi.org/10.1353/jda.2016.0134 https://scholar.google.com/scholar?hl=en&q=inancial%20portfolio%20approach%20to%20optimal%20tourist%20market%20mixes https://scholar.google.com/scholar?hl=en&q=inancial%20portfolio%20approach%20to%20optimal%20tourist%20market%20mixes http://dx.doi.org/10.1016/j.tourman.2007.09.003 https://scholar.google.com/scholar?hl=en&q=The%20exchange%20rate%20exposure%20of%20U.S.%20multinationals http://dx.doi.org/10.1086/296510 http://dx.doi.org/10.1016/j.tourman.2005.05.011 https://scholar.google.com/scholar?hl=en&q=Optimal%20foreign%20exchange%20risk%20hedging:%20A%20mean-variance%20portfolio%20approach https://scholar.google.com/scholar?hl=en&q=Optimal%20foreign%20exchange%20risk%20hedging:%20A%20mean-variance%20portfolio%20approach http://dx.doi.org/10.4236/tel.2013.31001 https://scholar.google.com/scholar?hl=en&q=Stationary%20vector%20autoregressive%20representation%20of%20error%20correction%20models http://dx.doi.org/10.4236/tel.2012.22027 https://scholar.google.com/scholar?hl=en&q=Management%20currency%20exposure:%20The%20case%20of%20Western%20mining https://scholar.google.com/scholar?hl=en&q=Management%20currency%20exposure:%20The%20case%20of%20Western%20mining http://dx.doi.org/10.1111/j.1745-6622.1990.tb00185.x https://scholar.google.com/scholar?hl=en&q=Portfolio%20selection https://scholar.google.com/scholar?hl=en&q=Beta,%20size,%20book-to-market%20equity%20and%20returns:%20A%20study%20based%20on%20UK%20data http://dx.doi.org/10.1016/j.mulfin.2006.12.003 https://scholar.google.com/scholar?hl=en&q=The%20contribution%20of%20tourism%20development%20to%20economic%20growth%20in%20the%20Korean%20economy http://dx.doi.org/10.1016/j.tourman.2003.09.014 https://scholar.google.com/scholar?hl=en&q=Explaining%20exchange%20rate%20risk%20in%20world%20stock%20markets:%20A%20panel%20approach http://dx.doi.org/10.1016/s0378-4266(01)00178-9 https://scholar.google.com/scholar?hl=en&q=Exchange%20rate%20volatility%20and%20manufacturing%20trade:%20Evidence%20from%20Africa http://dx.doi.org/10.1353/jda.2016.0125 https://scholar.google.com/scholar?hl=en&q=Exchange%20rate%20exposure%20of%20exporting%20and%20importing%20firms http://dx.doi.org/10.1111/j.1745-6622.2005.00047.x https://scholar.google.com/scholar?hl=en&q=Comparisons%20of%20riskiness%20as%20measured%20by%20the%20coefficient%20of%20variation https://scholar.google.com/scholar?hl=en&q=Comparisons%20of%20riskiness%20as%20measured%20by%20the%20coefficient%20of%20variation http://dx.doi.org/10.2307/251816 https://scholar.google.com/scholar?hl=en&q=Capital%20asset%20pricing:%20A%20theory%20of%20market%20equilibrium%20under%20conditions%20of%20risk https://scholar.google.com/scholar?hl=en&q=Capital%20asset%20pricing:%20A%20theory%20of%20market%20equilibrium%20under%20conditions%20of%20risk http://dx.doi.org/10.1111/j.1540-6261.1964.tb02865.x https://scholar.google.com/scholar?hl=en&q=Does%20risk%20management%20add%20value?%20A%20survey%20of%20the%20evidence http://dx.doi.org/10.1111/j.1745-6622.2005.00042.x https://scholar.google.com/scholar?hl=en&q=Forecasting%20tourism%20demand:%20A%20review%20of%20empirical%20research http://dx.doi.org/10.1016/0169-2070(95)00591-7