Environmental and Ecological Statistics

, Volume 22, Issue 4, pp 705–724 | Cite as

Inhomogeneous evolutionary MCMC for Bayesian optimal sequential environmental monitoring

  • Marco A. R. FerreiraEmail author


We develop a novel computational framework for Bayesian optimal sequential network design for environmental monitoring. This computational framework is based on inhomogeneous evolutionary Markov chain Monte Carlo, which combines ideas of genetic or evolutionary algorithms, Markov chain Monte Carlo, and inhomogenous Markov chains. Our framework allows optimality criteria with general utility functions that may include competing objectives, such as for example minimization of costs, minimization of the distance between true and estimated functions, and minimization of the prediction error. We illustrate our novel methodology with two applications to design of monitoring networks for ozone. The first application considers a one-time reduction of an existing network. The second application considers the design of a dynamic monitoring network where at each time point only a portion of the nodes of the network provide real time data.


Bayesian inference Evolutionary Monte Carlo  Kernel function estimation Optimal experimental design 



The work of Ferreira was supported in part by National Science Foundation Grant DMS-0907064. Part of this research was performed while Ferreira was visiting the Statistical and Applied Mathematical Sciences Institute (SAMSI). We gratefully acknowledge the constructive comments and suggestions made by two anonymous referees and the Associate Editor that led to a substantially improved article.


  1. Alves MB, Gamerman D, Ferreira MAR (2010) Transfer functions in dynamic generalized linear models. Stat Model 10:03–40CrossRefGoogle Scholar
  2. Amzal B, Bois FY, Parent E, Robert CP (2006) Bayesian-optimal design via interacting particle systems. J Am Stat Assoc 101:773–785CrossRefGoogle Scholar
  3. Berger J (1985) Statistical decision theory and Bayesian analysis. Springer, New YorkCrossRefGoogle Scholar
  4. Boer E, Dekkers A, Stein A (2002) Optimization of a monitoring network for sulfur dioxide. J Environ Qual 31(1):121–128CrossRefPubMedGoogle Scholar
  5. Carter CK, Kohn R (1994) On Gibbs sampling for state space models. Biometrika 81(3):541–553CrossRefGoogle Scholar
  6. Caselton WF, Husain T (1980) Hydrologic networks: information transmission. J Water Resour Plan Manag Div 106(2):503–520Google Scholar
  7. Caselton WF, Zidek JV (1984) Optimal monitoring network designs. Stat Prob Lett 2(4):223–227CrossRefGoogle Scholar
  8. Chaloner K, Verdinelli I (1995) Bayesian experimental design: a review. Stat Sci 10:273–304CrossRefGoogle Scholar
  9. Correia AW, Pope CA III, Dockery DW, Wang Y, Ezzati M, Dominici F (2013) Effect of air pollution control on life expectancy in the United States: an analysis of 545 US counties for the period from 2000 to 2007. Epidemiology 24(1):23–31PubMedCentralCrossRefPubMedGoogle Scholar
  10. DeGroot M (1970) Optimal statistical decisions. McGraw Hill, New YorkGoogle Scholar
  11. Diggle P, Lophaven S (2006) Bayesian geostatistical design. Scand J Stat 33(1):53–64CrossRefGoogle Scholar
  12. Dominici F, Peng RD, Zeger SL, White RH, Samet JM (2007) Particulate air pollution and mortality in the United States: did the risks change from 1987 to 2000? Am J Epidemiol 166(8):880–888CrossRefPubMedGoogle Scholar
  13. Doucet A, de Freitas N, Gordon N (2001) Sequential Monte Carlo methods in practice. Springer, New YorkCrossRefGoogle Scholar
  14. Drugan M, Thierens D (2004) Evolutionary Markov chain Monte Carlo. In: Liardet P, Collet P, Fonlupt C, Luton E, Shoenauer M (eds) Artificial evolution: 6th international conference, evolution artificielle, EA, 2003, Lecture Notes in Computer Science. Springer, Berlin, pp 63–76Google Scholar
  15. Ferreira MAR, Sanyal N (2014) Bayesian optimal sequential design for nonparametric regression via inhomogeneous evolutionary MCMC. Stat Methodol 18:131–141CrossRefGoogle Scholar
  16. Frühwirth-Schnatter S (1994) Data augmentation and dynamic linear models. J Time Ser Anal 15:183–202CrossRefGoogle Scholar
  17. Gamerman D, Lopes HF (2006) Markov chain Monte Carlo: stochastic simulation for Bayesian inference, 2nd edn. Chapman and Hall/CRC, Boca RatonGoogle Scholar
  18. Higdon DM (2002) Space and space–time modeling using process convolutions. In: Anderson CW, Barnett V, Chatwin PC, El-Shaarawi AH (eds) Quantitative methods for current environmental issues. Springer, New York, pp 37–54CrossRefGoogle Scholar
  19. Huerta G, Sansó B, Stroud JR (2004) A spatiotemporal model for Mexico City ozone levels. J R Stat Soc: Ser C 53(2):231–248CrossRefGoogle Scholar
  20. Jacquier E, Johannes M, Polson N (2007) MCMC maximum likelihood for latent state models. J Econom 127:615–640CrossRefGoogle Scholar
  21. Johansen AM, Doucet A, Davy M (2008) Particle methods for maximum likelihood estimation in latent variable models. Stat Comput 18:47–57CrossRefGoogle Scholar
  22. Li W, Han J (2010) Dynamic wireless sensor network parameters optimization adapting different node mobility. In: Proceedings of the IEEE aerospace conference, 2010, paper # 1170. IEEEGoogle Scholar
  23. Liu JS (2004) Monte Carlo strategies in scientific computing. Springer, New YorkCrossRefGoogle Scholar
  24. Müller P, Sansó B, Iorio MD (2004) Optimal Bayesian design by inhomogeneous Markov chain simulation. J Am Stat Assoc 99:788–798CrossRefGoogle Scholar
  25. Rappold AG, Gelfand AE, Holland DM (2008) Modelling mercury deposition through latent space–time processes. J R Stat Soc: Ser C 57(2):187–205CrossRefGoogle Scholar
  26. Robert CP, Casella G (2004) Monte Carlo statistical methods, 2nd edn. Springer, New YorkCrossRefGoogle Scholar
  27. Ruiz-Cárdenas R, Ferreira MAR, Schmidt AM (2010) Stochastic search algorithms for optimal monitoring network designs. Environmetrics 21:102–112Google Scholar
  28. Ruiz-Cárdenas R, Ferreira MAR, Schmidt AM (2012) Evolutionary Markov chain Monte Carlo algorithms for optimal monitoring network designs. Stat Methodol 9:185–194CrossRefGoogle Scholar
  29. Sinha A, Chandrakasan A (2001) Dynamic power management in wireless sensor networks. Des Test Comput IEEE 18(2):62–74CrossRefGoogle Scholar
  30. Song W-Z, Huang R, Xu M, Ma A, Shirazi B, LaHusen R (2009) Air-dropped sensor network for real-time high-fidelity volcano monitoring. In: Proceedings of the 7th international conference on mobile systems, applications, and services. ACM, pp 305–318Google Scholar
  31. Trujillo-Ventura A, Hugh Ellis J (1991) Multiobjective air pollution monitoring network design. Atmos Environ A Gen Top 25(2):469–479CrossRefGoogle Scholar
  32. West M, Harrison J (1997) Bayesian forecasting and dynamic models, 2nd edn. Springer, New YorkGoogle Scholar
  33. Wikle CK, Royle JA (2005) Dynamic design of ecological monitoring networks for non-Gaussian spatio-temporal data. Environmetrics 16:507–522CrossRefGoogle Scholar
  34. Zidek JV, Le ND, Liu Z (2012) Combining data and simulated data for space–time fields: application to ozone. Environ Ecol Stat 19(1):37–56CrossRefGoogle Scholar
  35. Zidek JV, Sun W, Le ND (2000) Designing and integrating composite networks for monitoring multivariate Gaussian pollution fields. J R Stat Soc: Ser C 49(1):63–79CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2015

Authors and Affiliations

  1. 1.Department of StatisticsVirginia TechBlacksburgUSA

Personalised recommendations