Consistency of the k-Nearest Neighbor Classifier for Spatially Dependent Data

Abstract

The purpose of this paper is to investigate the k-nearest neighbor classification rule for spatially dependent data. Some spatial mixing conditions are considered, and under such spatial structures, the well known k-nearest neighbor rule is suggested to classify spatial data. We established consistency and strong consistency of the classifier under mild assumptions. Our main results extend the consistency result in the i.i.d. case to the spatial case.

Keywords

Bayes rule / Spatial data / Training data / k-nearest neighbor rule / Mixing condition / Consistency

References

[1] Berbee, H.C.P.: Random walks with stationary increments and renewal theory. Math. Cent. Tracts. Amsterdam 58 (1979)

[2] Biau, G., Devroye, L.: Lectures on the Nearest Neighbor Method (2015)

[3] Bosq, D., Lecoutre, J.P.: Théorie de l’estimation fonctionnelle. Economica, Paris (1987)

[4] Cheng PE. Strong consistency of nearest neighbor regression function estimators. J. Multivar. Anal.. 1984, 15 63-72

[5] Devroye L, Györfi L, Krzyzak A, Lugosi G. On the strong universal consistency of nearest neighbor regression function estimates. Ann. Stat.. 1994, 22 1371-1385

[6] Devroye L, Györfi L, Lugosi G. A Probabilitic Theory of Pattern Recognition. 1996 New York: Springer

[7] Doukhan, P., Massart, P., Rio, E.: Invariance principles for absolutely regular empirical processes. Ann. Inst. H. Poincaré Probab. Statist. 31, 393–427 (1995)

[8] McDiarmid, C.: On the method of bounded differences. In: Surveys in Combinatorics, vol. 794. Cambridge University Press, Cambridge, pp. 261–283 (1989)

[9] Neaderhouser CC. Convergence of block spins defined on random fields. J. Statist. Phys.. 1980, 22 673-684

[10] Rio, E.: Théorie asymptotique des processus aléatoires faiblement dépendents. Mathématiques et Applications. Spriner, Berlin (2000)

[11] Rosenblatt M. A central limit theorem and a strong mixing condition. Ann. Stat.. 1977, 5 595-645

[12] Rosenblatt M. Stationary Sequences and Random Fields. 1985 Boston: Birkhauser

[13] Rozanov YA, Volkonskii VA. Some limit theorems for random functions. I. Teor. Veroyatn. Primen.. 1959, 4 186-207

[14] Stone, C.J.: Consistent nonparametric regression. Proc. Nat. Acad. Sci. USA 42, 43–47 (1956)

[15] Tran LT, Yakowitz S. Nearest neighbor estimators for random fields. J. Multivar. Anal.. 1993, 44 23-46

[16] Viennet, G.: Inequalities for absolutely sequence. Application to density estimation. Probab. Theory Relat. Fields 107, 467–492 (1967)

[17] Younso, A.: On nonparametric classification for weakly dependent functional processes. ESAIM: Probab. Stat. 21, 452–466 (2017)

[18] Younso A. On the consistency of a new kernel rule for spatially dependent data. Stat. Probab. Lett.. 2017, 131 64-71

[19] Younso A. On the consistency of kernel classification rule for functional random field. J. Soc. Française Stat.. 2018, 159 68-87

[20] Younso A. Nonparametric discrimination of areal functional data. Braz. J. Prob. Stat.. 2020, 34 12-126

[21] Younso A, Kanaya Z, Azhari N. Strong consistency of a kernel-based rule for spatially dependent data. Arab. J. Math. Sci.. 2019, 26 211-225

[22] Zhang X, Pan R, Guan G, Zhu X, Wang H. Network logistic regression model. Stat. Sin.. 2020, 30 673-693