In this paper, the proposed estimator for the bivariate intensity function for non-homogeneous Poisson process is based on multivariate continuous associated kernels. It is well known that, the serious problem inherent in this approach is that performance of the kernel estimator depends on the selection of a bandwidth parameter. To overcome the problem, we propose a Bayesian local approach to select the matrix of bandwidths considering the bivariate intensity function, and treating the bandwidths as a diagonal matrix of parameters with some prior distribution. To get an optimum parameter we have considered the selection of the local Bayesian bandwidth, which will then be compared, in terms of performance, to other approaches such as cross-validation method, Scott’s rule of thumbs, respectively. Note that the proposed method, in comparison in terms of minimizing the asymptotic mean integrated squared error, turn out favourably for the Bayesian method in examples studied.

Bayesian analysis, non-parametric inference, non-homogeneous Poisson process.

Received: November 22, 2020; Accepted: December 15, 2020; Published: March 18, 2021

How to cite this article: M. Badiane, P. Ngom and C. Manga, Bayesian selection of local bandwidth in non-homogeneous Poisson process kernel estimators for the intensity function, Far East Journal of Theoretical Statistics 61(2) (2021), 109-144. DOI: 10.17654/TS061020109

This Open Access Article is Licensed under Creative Commons Attribution 4.0 International License

References:

[1] A. Baddeley and R. Turner, Spatstat: an R package for analyzing spatial point patterns, J. Stat. Soft. 12(6) (2005), 1-42.
[2] C. D. Barr and F. P. Schoenberg, On the Voronoi estimator for the intensity of an inhomogeneous planar Poisson process, Biometrika 97(4) (2010), 977-984.
[3] M. Berman and P. Diggle, Estimating weighted integrals of the second-order intensity of a spatial point process, J. R. Stat. Soc. Ser. B. Stat. Methodol. 51 (1) (1989), 81-92.
[4] N. S. Belaid, N. S. Adjabi, C. C. Kokonendji and N. Zougab, Bayesian local bandwidth selector in multivariate associated kernel estimators for joint probability mass functions, Journal of Statistical Computation and Simulation 86 (2016a), 3667-3681.
[5] M. J. Brewer, A modelling approach for bandwidth selection in kernel density estimation, Proceedings of COMPSTAT Physica Verlag, Heidelberg, 1998, pp. 203-208.
[6] M. J. Brewer, A Bayesian model for local smoothing in kernel density estimation, Statistics and Computing 10 (2000), 299-309.
[7] M. M. Brooks and J. S. Marron, Asymptotic optimality of the least-squares cross-validation bandwidth for kernel estimates of intensity functions, Stochastic Process. Appl. 38(1) (1991), 157-165.
[8] R. Cao, Bootstrapping the mean integrated squared error, J. Multivariate Anal. 45(1) (1993), 137-160.
[9] R. Cao, A. Cuevas and W. González Manteiga, A comparative study of several smoothing methods in density estimation, Comput. Statist. Data Anal. 17(2) (1994), 153-176.
[10] A. Cowling, P. Hall and M. J. Phillips, Bootstrap confidence regions for the intensity of a Poisson point process, J. Amer. Statist. Assoc. 91(436) (1996), 1516-1524.
[11] Noel A. C. Cressie, Statistics for Spatial Data, (Revised), Wiley-Interscience, New York, 1993.
[12] O. Cronie and M. N. M. Van Lieshout, A non-model based approach to bandwidth selection for kernel estimators of spatial intensity functions, Biometrika 105 (2018), 455-462.
[13] L. Cucala, Espacements bidimensionnels et données entachées dérreurs dans l’analyse des processus ponctuels spatiaux, Ph. D. Thesis, Université des Sciences Sociales, Toulouse I, 2006.
[14] L. Cucala, Intensity estimation for spatial point processes observed with noise, Scand. J. Stat. 35(2) (2008), 322-334.
[15] M. S. De Lima and G. S. Atuncar, Unconstrained pilot selectors for smoothed cross-validation, Journal of Nonparametric Statistics 23 (2010), 137-148.
[16] P. Diggle, A kernel method for smoothing point process data, J. R. Stat. Soc. Ser. C. Appl. Stat. 34(2) (1985), 138-147.
[17] P. J. Diggle, Statistical Analysis of Spatial Point Patterns, 2nd ed., A Hodder Arnold Publication, Oxford, UK, 2003.
[18] P. Diggle and J. S. Marron, Equivalence of smoothing parameter selectors in density and intensity estimation, J. Amer. Statist. Assoc. 83 (1988), 793-800.
[19] T. Duong, ks: Kernel Smoothing, R package version 1.8.13, 2013.
[20] T. Duong and M. Hazelton, Plug-in bandwidth matrices for bivariate kernel density estimation, J. Nonparametr. Stat. 15(1) (2003), 17-30.
[21] T. Duong and M. L. Hazelton, Cross-validation bandwidth matrices for multivariate kernel density estimation, Scand. J. Stat. 32(3) (2005), 485-506.
[22] I. Fuentes-Santos, W. González-Manteiga and J. Mateu, Consistent smooth bootstrap kernel intensity estimation for inhomogeneous spatial Poisson point processes, Scand. J. Stat. 43(2) (2015), 416-435.
[23] W. González-Manteiga, R. Cao and J. S. Marron, Bootstrap selection of the smoothing parameter in nonparametric hazard rate estimation, J. Amer. Statist. Assoc. 91(435) (1996), 1130-1140.
[24] Y. Guan, A goodness-of-fit test for inhomogeneous spatial Poisson processes, Biometrika 95(4) (2008a), 831-845.
[25] Y. Guan, On consistent nonparametric intensity estimation for inhomogeneous spatial point processes, J. Amer. Statist. Assoc. 103(483) (2008b), 1238-1247.
[26] Y. Guan and J. M. Loh, A thinned block bootstrap variance estimation procedure for inhomogeneous spatial point patterns, J. Amer. Statist. Assoc. 102(480) (2007), 1377-1386.
[27] A. K. Gangopadhyay and K. N. Cheung, Bayesian approach to the choice of smoothing parameter in kernel density estimation, J. Nonparametric Stat. 14 (2002), 655-664.
[28] D. J. Illian, P. A. Penttinen, D. H. Stoyan and D. Stoyan, Statistical Analysis and Modelling of Spatial Point Patterns, John Wiley & Sons, Chichester, UK, 2008.
[29] K. B. Kulasekera and W. J. Padgett, Bayes bandwidth selection in kernel density estimation with censored data, Journal of Nonparametric Statistics 18 (2006), 129-143.
[30] C. N. Kuruwita, K. B. Kulasekera and W. J. Padgett, Density estimation using asymmetric kernels and Bayes bandwidths with censored data, Journal of Statistical Planning and Inference 140 (2010), 1765-1774.
[31] J. M. Loh and W. Jang, Estimating a cosmological mass bias parameter with bootstrap bandwidth selection, J. R. Stat. Soc. Ser. C. Appl. Stat. 59(5) (2010), 761-779.
[32] P. H. Lo, An iterative plug-in algorithm for optimal bandwidth selection in kernel intensity estimation for spatial data, Ph.D. Thesis, Technical University of Kaiserslautern, 2017.
[33] J. S. Marron and M. P. Wand, Exact mean integrated squared error, Ann. Statist. 20(2) (1992), 712-736.
[34] D. Scott, Multivariate Density Estimation, Theory, Practice and Visualization, Wiley, New York, 1992.
[35] B. Silverman, Density Estimation for Statistics and Data Analysis (Monographs on Statistics and Applied Probability), Chapman & Hall/CRC, New York, 1986.
[36] S. M. Somé, Estimation non-paramétrique par noyaux associés multivariés et applications, Thèse de Doctorat, Université de Bourgogne Franche-Comté, France, 2015.
[37] C. C. Taylor, Bootstrap choice of the smoothing parameter in kernel density estimation, Biometrika 76(4) (1989), 705.
[38] R. P. Waagepetersen, An estimating function approach to inference for inhomogeneous Neyman-Scott processes, Biometrics 63(1) (2007), 252-258.
[39] M. P. Wand and M. C. Jones, Multivariate plug-in bandwidth selection, Computational Statistics 9 (1994), 97-116.
[40] M. P. Wand, Error analysis for general multivariate kernel estimators, J. Nonparametr. Stat. 2(1) (1992), 1-15.
[41] X. Zhang, M. L. King and R. J. Hyndman, A Bayesian approach to bandwidth selection for multivariate kernel density estimation, Computational Statistics and Data Analysis 50 (2006), 3009-3031.
[42] Y. Ziane, S. Adjabi and N. Zougab, Adaptive Bayesian bandwidth selection in asymmetric kernel density estimation for nonnegative heavy-tailed data, J. Appl. Stat. 42 (2015), 1645-1658.
[43] N. Zougab, S. Adjabi and C. C. Kokonendji, Binomial kernel and Bayes local bandwidth in discrete functions estimation, J. Nonparametric Stat. 24 (2012), 783-795.
[44] N. Zougab, S. Adjabi and C. C. Kokonendji, Adaptive smoothing in associated kernel discrete functions estimation using Bayesian approach, J. Stat. Comput. Simul. 83 (2013), 2219-2231.
[45] N. Zougab, S. Adjabi and C. C. Kokonendji, Bayesian estimation of adaptive bandwidth matrices in multivariate kernel density estimation, Computational Statistics and Data Analysis 75 (2014), 28-38.