Density estimation over spatio-temporal data streams

Abstract In the last few years, data can be collected extremely easily in many scientific research fields. This became possible by the recent technological advances that have made online monitoring possible. In such situations, if real time or online estimations are expected, the usual nonparametric techniques rapidly require a lot of time to be computed and therefore become useless in practice. Adaptative counterparts of the classical kernel density estimators, that can be updated extremely easily when a new set of observations is available are investigated, for spatio-temporal processes with weak dependence structures. Mean square, uniform almost sure convergences and a central limit result are obtained under general and easily verifiable conditions. The efficiency of the considered estimators is evaluated through simulations and a real data application. The results show that the proposed method works well within the framework of a spatio-temporal data stream.

[1]  L. Tran,et al.  Kernel density estimation for random fields (density estimation for random fields) , 1997 .

[2]  H. Robbins.  A Stochastic Approximation Method , 1951 .

[3]  Leila Hamdad,et al.  A kernel spatial density estimation allowing for the analysis of spatial clustering. Application to Monsoon Asia Drought Atlas data , 2014, Stochastic Environmental Research and Risk Assessment.

[4]  L. Tran.  Kernel density estimation on random fields , 1990 .

[5]  Aboubacar Amiri.  Sur une famille paramétrique d'estimateurs séquentiels de la densité pour un processus fortement mélangeant , 2009 .

[6]  Edward J. Wegman,et al.  Remarks on Some Recursive Estimators of a Probability Density , 1979 .

[7]  Ping-Shi Wu,et al.  Real-Time Density and Mode Estimation With Application to Time-Dynamic Mode Tracking , 2006 .

[8]  G. M. Robison.  Divergent double sequences and series , 1926 .

[9]  D. Shin,et al.  Kernel estimators of mode under $$\psi $$ψ-weak dependence , 2016 .

[10]  Jinde Wang,et al.  Estimation of the trend function for spatio-temporal models , 2009 .

[11]  Li Wei,et al.  M-kernel merging: towards density estimation over data streams , 2003, Eighth International Conference on Database Systems for Advanced Applications, 2003. (DASFAA 2003). Proceedings..

[12]  P. Robinson,et al.  Asymptotic theory for nonparametric regression with spatial data , 2011 .

[13]  C. Francq,et al.  Kernel regression estimation for random fields , 2007 .

[14]  I. Ibragimov,et al.  Some Limit Theorems for Stationary Processes , 1962 .

[15]  Marc Hallin,et al.  Kernel density estimation for spatial processes: the L 1 theory , 2004 .

[16]  Xiaohong Chen,et al.  Recursive Nonparametric Estimation for Time Series , 2014, IEEE Transactions on Information Theory.

[17]  Nazgul Jenish.  Nonparametric spatial regression under near-epoch dependence , 2012 .

[18]  Marc Hallin,et al.  Density estimation for spatial linear processes , 2001 .

[19]  On the Estimation of the Density of a Directional Data Stream , 2017 .

[20]  Jiti Gao,et al.  Estimation in semiparametric spatial regression , 2003 .

[21]  Timothy G. Conley.  GMM estimation with cross sectional dependence , 1999 .

[22]  D. Tjøstheim,et al.  Nonparametric Estimation of Probability Density Functions for Irregularly Observed Spatial Data , 2014 .

[23]  J. Dedecker,et al.  New dependence coefficients. Examples and applications to statistics , 2005 .

[24]  S. Dabo‐Niang,et al.  KERNEL REGRESSION ESTIMATION FOR SPATIAL FUNCTIONAL RANDOM VARIABLES , 2010 .

[25]  Tilman M. Davies,et al.  Adaptive kernel estimation of spatial relative risk , 2010, Statistics in medicine.

[26]  M. Rosenblatt.  A CENTRAL LIMIT THEOREM AND A STRONG MIXING CONDITION. , 1956, Proceedings of the National Academy of Sciences of the United States of America.

[27]  Lanh Tat Tran.  Recursive density estimation under dependence , 1989, IEEE Trans. Inf. Theory.

[28]  P. Doukhan,et al.  A new weak dependence condition and applications to moment inequalities , 1999 .

[29]  Haibo He,et al.  SOMKE: Kernel Density Estimation Over Data Streams by Sequences of Self-Organizing Maps , 2012, IEEE Transactions on Neural Networks and Learning Systems.

[30]  Hisao Ishibuchi,et al.  Dm-KDE: dynamical kernel density estimation by sequences of KDE estimators with fixed number of components over data streams , 2014, Frontiers of Computer Science.

[31]  P. Doukhan,et al.  Functional Estimation of a Density Under a New Weak Dependence Condition , 2001 .

[32]  Kernel density estimation for stationary random fields , 2011, 1109.2694.

[33]  Keming Yu,et al.  Local linear spatial quantile regression , 2009 .

[34]  Dag Tjøstheim,et al.  Adaptively varying‐coefficient spatiotemporal models , 2009 .

[35]  Clémentine Prieur,et al.  Coupling for τ-Dependent Sequences and Applications , 2004 .

[36]  M. Hallin,et al.  Local linear spatial regression , 2004, math/0508597.

[37]  S. Louhichi,et al.  Recursive kernel estimation of the density under η-weak dependence , 2014 .

[38]  I. Fazekas,et al.  Asymptotic Normality of Kernel Type Density Estimators for Random Fields , 2006 .

[39]  Han-ying Liang,et al.  Asymptotic normality of recursive density estimates under some dependence assumptions , 2004 .

[40]  Bo Huang,et al.  Prediction for spatio-temporal models with autoregression in errors , 2012 .

[41]  Jun Zhu,et al.  Estimation for semiparametric nonlinear regression of irregularly located spatial time-series data , 2017 .

[42]  G. Biau,et al.  Nonparametric Spatial Prediction , 2004 .

[43]  Noel A Cressie,et al.  Statistics for Spatial Data. , 1992 .

[44]  Elias Masry,et al.  Recursive probability density estimation for weakly dependent stationary processes , 1986, IEEE Trans. Inf. Theory.

[45]  S. Dabo‐Niang,et al.  Nonparametric recursive density estimation for spatial data , 2016 .