Advertisement

Efficient Sequential Monte Carlo Algorithms for Integrated Population Models

  • Axel FinkeEmail author
  • Ruth King
  • Alexandros Beskos
  • Petros Dellaportas
Article
  • 36 Downloads

Abstract

In statistical ecology, state-space models are commonly used to represent the biological mechanisms by which population counts—often subdivided according to characteristics such as age group, gender or breeding status—evolve over time. As the counts are only noisily or partially observed, they are typically not sufficiently informative about demographic parameters of interest and must be combined with additional ecological observations within an integrated data analysis. Fitting integrated models can be challenging, especially if the constituent state-space model is nonlinear/non-Gaussian. We first propose an efficient particle Markov chain Monte Carlo algorithm to estimate demographic parameters without a need for linear or Gaussian approximations. We then incorporate this algorithm into a sequential Monte Carlo sampler to perform model comparison. We also exploit the integrated model structure to enhance the efficiency of both algorithms. The methods are demonstrated on two real data sets: little owls and grey herons. For the owls, we find that the data do not support an ecological hypothesis found in the literature. For the herons, our methodology highlights the limitations of existing models which we address through a novel regime-switching model. Supplementary materials accompanying this paper appear online.

Keywords

Bayesian inference Capture–recapture Integrated population models Model comparison Sequential Monte Carlo State-space models 

Notes

Acknowledgements

This work was supported by The Alan Turing Institute under the EPSRC Grant EP/N510129/1. AB and AF were funded by a Leverhulme Trust Prize.

Supplementary material

13253_2018_349_MOESM1_ESM.pdf (345 kb)
Web Appendices and Figures, referenced in Sections 3--6, are available at [link to supplementary pdf file goes here]. All data and C++/R code necessary for reproducing the results can be found at https://github.com/AxelFinke/monte-carlo-rcpp. (345KB)

References

  1. Abadi, F., Gimenez, O., Arlettaz, R., and Schaub, M. (2010a). An assessment of integrated population models: Bias, accuracy, and violation of the assumption of independence. Ecology, 91(1):7–14.Google Scholar
  2. Abadi, F., Gimenez, O., Ullrich, B., Arlettaz, R., and Schaub, M. (2010b). Estimation of immigration rate using integrated population models. Journal of Applied Ecology, 47(2):393–400.Google Scholar
  3. Andrieu, C., Doucet, A., and Holenstein, R. (2010). Particle Markov chain Monte Carlo methods. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 72(3):269–342. With discussion.MathSciNetzbMATHGoogle Scholar
  4. Bernardo, J. M. and Smith, A. F. M. (2009). Bayesian Theory. Wiley.Google Scholar
  5. Besbeas, P., Borysiewicz, R. S., and Morgan, B. J. T. (2009). Completing the Ecological Jigsaw. In Modeling Demographic Processes in Marked Populations, pages 513–539. Springer.Google Scholar
  6. Besbeas, P., Freeman, S. N., Morgan, B. J. T., and Catchpole, E. A. (2002). Integrating mark-recapture-recovery and census data to estimate animal abundance and demographic parameters. Biometrics, 58(3):540–547.Google Scholar
  7. Besbeas, P. and Morgan, B. J. T. (2012). A threshold model for heron productivity. Journal of Agricultural, Biological, and Environmental Statistics, 17(1):128–141.MathSciNetzbMATHGoogle Scholar
  8. Breed, G., Costa, D., Jonsen, I., Robinson, P., and Mills-Flemming, J. (2012). State-space methods for more completely capturing behavioral dynamics from animal tracks. Ecological Modelling, 235:49–58.Google Scholar
  9. Brooks, S. P., King, R., and Morgan, B. J. T. (2004). A Bayesian approach to combining animal abundance and demographic data. Animal Biodiversity and Conservation, 27(1):515–529.Google Scholar
  10. Buckland, S. T., Newman, K. B., Fernandez, C., Thomas, L., and Harwood, J. (2007). Embedding population dynamics models in inference. Statistical Science, 22(1):44–58.MathSciNetzbMATHGoogle Scholar
  11. Carpenter, B., Gelman, A., Hoffman, M. D., Lee, D., Goodrich, B., Betancourt, M., Brubaker, M., Guo, J., Li, P., and Riddell, A. (2017). Stan: A probabilistic programming language. Journal of Statistical Software, 76(1).Google Scholar
  12. Chopin, N. (2002). A sequential particle filter method for static models. Biometrika, 89(3):539–552.MathSciNetzbMATHGoogle Scholar
  13. Chopin, N., Jacob, P. E., and Papaspiliopoulos, O. (2013). \(\text{ SMC }^{2}\): An efficient algorithm for sequential analysis of state space models. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 75(3):397–426.MathSciNetGoogle Scholar
  14. Christen, J. A. and Fox, C. (2005). Markov chain Monte Carlo using an approximation. Journal of Computational and Graphical Statistics, 14(4):795–810.MathSciNetGoogle Scholar
  15. de Valpine, P., Turek, D., Paciorek, C. J., Anderson-Bergman, C., Lang, D. T., and Bodik, R. (2017). Programming with models: Writing statistical algorithms for general model structures with NIMBLE. Journal of Computational and Graphical Statistics, 26(2):403–413.MathSciNetGoogle Scholar
  16. Del Moral, P. (1996). Nonlinear filtering: Interacting particle solution. Markov Processes and Related Fields, 2(4):555–580.MathSciNetzbMATHGoogle Scholar
  17. Del Moral, P., Doucet, A., and Jasra, A. (2006). Sequential Monte Carlo samplers. Journal of the Royal Statistical Society: Series B, 68(3):411–436.MathSciNetzbMATHGoogle Scholar
  18. Doucet, A. and Johansen, A. M. (2011). A tutorial on particle filtering and smoothing: Fifteen years later. In Crisan, D. and Rozovskii, B., editors, The Oxford Handbook of Nonlinear Filtering, Oxford Handbooks, chapter 24, pages 656–704. Oxford University Press.Google Scholar
  19. Drovandi, C. C. and McCutchan, R. A. (2016). Alive \(\text{ SMC }^{2}\): Bayesian model selection for low-count time series models with intractable likelihoods. Biometrics, 72(2):344–353.MathSciNetzbMATHGoogle Scholar
  20. Duan, J.-C. and Fulop, A. (2015). Density-tempered marginalized sequential Monte Carlo samplers. Journal of Business & Economic Statistics, 33(2):192–202.MathSciNetGoogle Scholar
  21. Dupuis, J. A. (1995). Bayesian estimation of movement and survival probabilities from capture-recapture data. Biometrika, 82(4):761–772.MathSciNetzbMATHGoogle Scholar
  22. Gilks, W. R., Thomas, A., and Spiegelhalter, D. J. (1994). A language and program for complex Bayesian modelling. Journal of the Royal Statistical Society. Series D (The Statistician) , 43(1):169–177.Google Scholar
  23. Golightly, A., Henderson, D. A., and Sherlock, C. (2015). Delayed acceptance particle MCMC for exact inference in stochastic kinetic models. Statistics and Computing, 25(5):1039–1055.MathSciNetzbMATHGoogle Scholar
  24. Green, P. J. (1995). Reversible jump Markov chain Monte Carlo computation and Bayesian model determination. Biometrika, 82(4):711–732.MathSciNetzbMATHGoogle Scholar
  25. Kalman, R. E. (1960). A new approach to linear filtering and prediction problems. Journal of Basic Engineering, 82(1):35–45.Google Scholar
  26. Kass, R. E. and Raftery, A. E. (1995). Bayes factors. Journal of the American Statistical Association, 90(430):773–795.MathSciNetzbMATHGoogle Scholar
  27. King, R. (2011). Statistical Ecology. In Brooks, S., Gelman, A., Jones, G., and Meng, X.-L., editors, Handbook of Markov Chain Monte Carlo, chapter 17, pages 410–447. CRC Press.Google Scholar
  28. King, R. (2012). A review of Bayesian state-space modelling of capture-recapture-recovery data. Interface Focus, 2:190–204.Google Scholar
  29. King, R. (2014). Statistical ecology. Annual Review of Statistics and its Application, 1(1):401–426.Google Scholar
  30. King, R., Brooks, S., Mazzetta, C., Freeman, S., and Morgan, B. (2008). Identifying and diagnosing population declines: A Bayesian assessment of lapwings in the UK. Journal of Royal Statistical Society: Series C, 57(5):609–632.MathSciNetGoogle Scholar
  31. Knape, J. and de Valpine, P. (2012). Fitting complex population models by combining particle filters with Markov chain Monte Carlo. Ecology, 93(2):256–263.Google Scholar
  32. Lindley, D. V. (1957). A statistical paradox. Biometrika, 44(1/2):187–192.zbMATHGoogle Scholar
  33. McClintock, B. T., King, R., Thomas, L., Matthiopoulos, J., McConnell, B. J., and Morales, J. M. (2012). A general discrete-time modeling framework for animal movement using multi-state random walks. Ecological Monographs, 82(3):335–349.Google Scholar
  34. McCrea, R. S. and Morgan, B. J. T. (2014). Analysis of Capture-Recapture Data. CRC Press.Google Scholar
  35. Millar, R. B. and Meyer, R. (2000). Non-linear state space modelling of fisheries biomass dynamics by using Metropolis-Hastings within-Gibbs sampling. Journal of the Royal Statistical Society: Series C, 49(3):327–342.MathSciNetzbMATHGoogle Scholar
  36. Morales, J., Haydon, D., Frair, J., Holsiner, K., and Fryxell, J. (2004). Extracting more out of relocation data: Building movement models as mixtures of random walks. Ecology, 85(9):2436–2445.Google Scholar
  37. Newman, K. B. (1998). State-space modelling of animal movement and mortality with application to salmon. Biometrics, 54:1290–1314.zbMATHGoogle Scholar
  38. Newman, K. B., Buckland, S. T., Morgan, B. J. T., King, R., Borchers, D. L., Cole, D., Besbeas, P. T., Gimenez, O., and Thomas, L. (2014). Modelling Population Dynamics: Model Formulation, Fitting and Assessment using State-space Methods. Springer.Google Scholar
  39. Nishimura, A., Dunson, D., and Lu, J. (2017). Discontinuous Hamiltonian Monte Carlo for models with discrete parameters and discontinuous likelihoods. ArXiv e-prints arXiv:1705.08510.
  40. Parslow, J., Cressie, N., Campbell, E. P., Jones, E., and Murray, L. (2013). Bayesian learning and predictability in a stochastic nonlinear dynamical model. Ecological Applications, 23(4):679–698.Google Scholar
  41. Peters, G. W., Hosack, G. R., and Hayes, K. R. (2010). Ecological non-linear state space model selection via adaptive particle Markov chain Monte Carlo (AdPMCMC). ArXiv e-prints arXiv:1005.2238.
  42. Plummer, M. (2003). JAGS: A program for analysis of Bayesian graphical models using Gibbs sampling. In Proceedings of the 3rd International Workshop on Distributed Statistical Computing. Vienna, Austria.Google Scholar
  43. Pooley, C. and Marion, G. (2018). Bayesian model evidence as a practical alternative to deviance information criterion. Royal Society Open Science, 5(3):171519.MathSciNetGoogle Scholar
  44. Royle, J. A. (2008). Modeling individual effects in the Cormack-Jolly-Seber model: A state-space formulation. Biometrics, 64(2):364–370.MathSciNetzbMATHGoogle Scholar
  45. Schaub, M., Ullrich, B., Knötzsch, G., Albrecht, P., and Meisser, C. (2006). Local population dynamics and the impact of scale and isolation: A study on different little owl populations. Oikos, 115(3):389–400.Google Scholar
  46. Sherlock, C., Thiery, A., and Golightly, A. (2015). Efficiency of delayed-acceptance random walk Metropolis algorithms. ArXiv e-prints arXiv:1506.08155.
  47. Spiegelhalter, D. J., Best, N. G., Carlin, B. P., and Van Der Linde, A. (2002). Bayesian measures of model complexity and fit. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 64(4):583–639.MathSciNetzbMATHGoogle Scholar
  48. Zhou, Y., Johansen, A. M., and Aston, J. A. D. (2016). Towards automatic model comparison: An adaptive sequential Monte Carlo approach. Journal of Computational and Graphical Statistics, 25(3):701–726.MathSciNetGoogle Scholar

Copyright information

© International Biometric Society 2019

Authors and Affiliations

  1. 1.Department of Statistics and Applied ProbabilityNational University of SingaporeSingaporeSingapore
  2. 2.Department of Statistical ScienceUniversity College LondonLondonUK
  3. 3.School of MathematicsUniversity of EdinburghEdinburghUK
  4. 4.The Alan Turing InstituteLondonUK
  5. 5.Department of StatisticsAthens University of Economics and BusinessAthensGreece

Personalised recommendations