Asymptotic analysis of statistical estimators of parameters for binomial conditionally autoregressive model of spatio-temporal data

Authors

  • Maryna K. Dauhaliova Belarusian State University, Niezaliežnasci Avenue, 4, 220030, Minsk, Belarus
  • Yuriy S. Kharin Research Institute for Applied Problems of Mathematics and Informatics, Belarusian State University, 4 Niezaliežnasci Avenue, Minsk 220030, Belarus; Belarusian State University, 4 Niezaliežnasci Avenue, Minsk 220030, Belarus

Keywords:

spatio-temporal data, inhomogeneous Markov chain, ergodic principle, maximum likelihood estimator, consistency of estimator, asymptotic normality

Abstract

The binomial conditionally autoregressive model of discrete spatio-temporal data is considered in this paper. This model is a multidimensional inhomogeneous Markov chain with a finite state space. Conditions, under which the binomial conditionally autoregressive model satisfies the ergodic principle, are found in case when exogenous factors depend on time. The maximum likelihood approach is used for statistical estimation of model parameters. It is proved that the constructed maximum likelihood estimators are consistent and asymptotically normal distributed for any bounded values of the model parameters and any bounded values of the exogenous factor in case of statistical identifiability of model parameters. Results of computer experiments on simulated data illustrate consistency of maximum likelihood estimators. 

Author Biographies

  • Maryna K. Dauhaliova, Belarusian State University, Niezaliežnasci Avenue, 4, 220030, Minsk, Belarus

    junior researcher

  • Yuriy S. Kharin, Research Institute for Applied Problems of Mathematics and Informatics, Belarusian State University, 4 Niezaliežnasci Avenue, Minsk 220030, Belarus; Belarusian State University, 4 Niezaliežnasci Avenue, Minsk 220030, Belarus

    corresponding member of the National Academy of Sciences of Belarus, doctor of science (physics and ma the matics); director, Research Institute for Applied Problems of Mathematics and Informatics, Belarusian State University; professor at the department of mathematical modeling and data analysis, faculty of applied mathematics and informatics, Belarusian State University

References

  1. Kharin Yu. Robustness in Statistical Forecasting. New York: Springer; 2013. DOI: 10.1007/978-3-319-00840-0.
  2. Kharin Yu, Zhurak M. Statistical analysis of spatio-temporal data based on Poisson model. Informatica. 2015;26(1):67–87. DOI: 10.15388/Informatica.2015.39.
  3. Kharin Yu, Zhurak M. Statistical forecasting based on binomial conditional autoregressive model of spatio-temporal data. Pliska Studia Mathematica. 2017;27:23–36.
  4. Gelfand AE. Handbook of Spatial Statistics. [USA]: Taylor and Francis Group, LLC; 2010.
  5. Boulieri A, Liverani S, Hoogh K, Blangiardo M. A space-time multivariate Bayesian model to analyse road traffic accidents by severity. Journal of the Royal Statistical Society. Series A. 2017;180(1):119–139. DOI: 10.1111/rssa.12178.
  6. Rushworth A, Lee D, Sarran C. An adaptive spatiotemporal smoothing model for estimating trends and step changes in disease risk. Journal of the Royal Statistical Society. Series C. 2017;66(1):141–157. DOI: 10.1111/rssc.12155.
  7. Amek N, Bayoh N, Hamel M, Lindblade KA, Gimnig J, Laserson KF, et al. Spatio-temporal modeling of sparse geosta‑ tistical malaria sporozoite rate data using a zero inflated binomial model. Spatial and Spatio-temporal Journal of Epidemiology. 2011;2(4):283–290. DOI: 10.1016/j.sste.2011.08.001.
  8. Kharin YuS, Zhurak MK. Binomial conditional autoregressive model of the space-time data and its probabilistic and statistical analysis. Doklady of the National Academy of Sciences of Belarus. 2015;59(6):5–12. Russian.
  9. Kharin YuS, Zhurak MK. Asymptotic analysis of maximum likelihood estimators for parameters of binomial conditionally autoregressive model of spatio-temporal data. Proceedings of the National Academy of Sciences. Physics and Mathematics Series. 2016;1:36 – 45. DOI: 10.29235/1561-2430-2016-0-1-36-45. Russian.
  10. Sarymsakov TA, Mustafin HA. On an ergodic theorem for inhomogeneous Markov chains. Trudy Sredneaziatskogo gosudarstvennogo universiteta im. V. I. Lenina. 1957;74:1–37. Russian.
  11. Lancaster P. Teoriya matrits [The theory of matrices]. Moscow: Nauka; 1973. Russian.
  12. Kharin YuS, Zuev NM, Zuk EE. Teoriya veroyatnostei, matematicheskaya i prikladnaya statistika [Probability theory, mathe‑ matical and applied statistics]. Minsk: Belarusian State University; 2011. Russian.
  13. Bernshtein SN. Teoriya veroyatnostei [Probability theory]. Moscow: Gosizdatel’stvo; 1927. Russian.
  14. Borovkov AA. Matematicheskaya statistika. Otsenka parametrov. Proverka gipotez [Math statistics. Evaluation of parameters. Hypotheses testing]. Moscow: Nauka; 1984. Russian.
  15. Dobrushin RL. Central limit theorem for nonstationary Markov Chains. I. Teoriya veroyatnostei i ee primeneniya. 1956;1(1): 72–89. Russian.

Downloads

Published

2019-01-19

Issue

Section

Probability Theory and Mathematical Statistics

How to Cite

[1]
Dauhaliova, M.K. and Kharin, Y.S. 2019. Asymptotic analysis of statistical estimators of parameters for binomial conditionally autoregressive model of spatio-temporal data. Journal of the Belarusian State University. Mathematics and Informatics. 2 (Jan. 2019), 47–57.