Layout 1 ISDS Annual Conference Proceedings 2012. This is an Open Access article distributed under the terms of the Creative Commons Attribution- Noncommercial 3.0 Unported License (http://creativecommons.org/licenses/by-nc/3.0/), permitting all non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. ISDS 2012 Conference Abstracts Tau-leaped Particle Learning Jarad Niemi*1 and Michael Ludkovski2 1Iowa State University, Ames, IA, USA; 2University of California, Santa Barbara, Santa Barbara, CA, USA Objective Develop fast sequential Bayesian inference for disease outbreak counts. Introduction Development of effective policy interventions to stem disease out- breaks requires knowledge of the current state of affairs, e.g. how many individuals are currently infected, a strain’s virulence, etc, as well as our uncertainty of these values. A Bayesian inferential ap- proach provides this information, but at a computational expense. We develop a sequential Bayesian approach based on an epidemiological compartment model and noisy count observations of the transitions between compartments. Methods For simplicity, consider an SIR epidemiological compartment model where compartments exist for susceptible, infected, and re- covered individuals. Transitions between compartments occur in dis- crete time with transitions numbers given by Poisson random variables, the tau-leaping approximation, whose means depend on the current compartment occupancy and some unknown fixed pa- rameters, e.g. virulence. Binomial observations, with possible un- known sampling proportion, are made on these transitions. The standard sequential Bayesian updating methodology is se- quential Monte Carlo (SMC), a.k.a. particle filtering. The original bootstrap filter is effective when the system has no fixed parameters, but exhibits marked degeneracy otherwise [1]. An approach based on resampling the fixed parameters from a kernel density estimate pro- vides a generic approach with less degeneracy [2]. We build methodology based on a particle learning framework [3]. In this framework, each particle carries a set of parameter-specific sufficient statistics and samples parameter values whenever neces- sary. In addition, the methodology promotes a resample-move ap- proach based on the predictive likelihood that reduces degeneracy in the first place. An improvement on the particle learning framework in this model is that some fixed parameters can be integrated out of the predictive likelihood. This Rao-Blackwellization provides an SMC methodol- ogy with reduced Monte Carlo variance. Results For a fixed number of particles or computational expense, we show improvements in accuracy relative to the kernel density approach and an alternative approach based on sufficient statistics [4] where com- pared with a gold-standard Markov chain Monte Carlo analysis. Conclusions Many surveillance systems collect counts of adverse events related to some disease. These counts are expected to be a fraction of the true underlying disease extent. The methodology developed here allows a fully Bayesian analysis that uncovers the true number of infected in- dividuals as well as disease virulence based on these count data. This statistical approach can be combined with an optimal policy map to help public health officials react effectively to initial disease reports. Keywords surveillance; Bayesian; sequential Monte Carlo; particle learning References [1] Gordon, Salmond, and Smith. Novel approach to nonlinear/non- Gaussian Bayesian state estimation. IEE Proceedings Part F: Com- munications, Radar and Signal Processing. 140(2): 107-113 (1993). [2] Liu and West. Combined parameter and state estimation in simula- tion-based filtering. Doucet, De Freitas, and Gordon, ed. Sequential Monte Carlo Methods in Practice. Springer-Verlag, New York. 197— 217 (2001). [3] Carvalho, Johannes, Lopes, and Polson. Particle learning and smooth- ing. Statistical Science. 25(1): 88—106 (2010). [4] Storvik. Particle filters in state space models with the presence of un- known static parameters. IEEE Transactions on Signal Processing. 50(2): 281—289 (2002). *Jarad Niemi E-mail: niemi@iastate.edu Online Journal of Public Health Informatics * ISSN 1947-2579 * http://ojphi.org * 5(1):e8, 2013