Local linear spatial regression
Hallin, Marc ; Lu, Zudi ; Tran, Lanh T.
Ann. Statist., Tome 32 (2004) no. 1, p. 2469-2500 / Harvested from Project Euclid
A local linear kernel estimator of the regression function x↦g(x):=E[Yi|Xi=x], x∈ℝd, of a stationary (d+1)-dimensional spatial process {(Y i,Xi),i∈ℤN} observed over a rectangular domain of the form ℐn:={i=(i1,…,iN)∈ℤN|1≤ik≤nk,k=1,…,N}, n=(n1,…,nN)∈ℤN, is proposed and investigated. Under mild regularity assumptions, asymptotic normality of the estimators of g(x) and its derivatives is established. Appropriate choices of the bandwidths are proposed. The spatial process is assumed to satisfy some very general mixing conditions, generalizing classical time-series strong mixing concepts. The size of the rectangular domain ℐn is allowed to tend to infinity at different rates depending on the direction in ℤN.
Publié le : 2004-12-14
Classification:  Mixing random field,  local linear kernel estimate,  spatial regression,  asymptotic normality,  62G05,  60J25,  62J02
@article{1107794876,
     author = {Hallin, Marc and Lu, Zudi and Tran, Lanh T.},
     title = {Local linear spatial regression},
     journal = {Ann. Statist.},
     volume = {32},
     number = {1},
     year = {2004},
     pages = { 2469-2500},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1107794876}
}
Hallin, Marc; Lu, Zudi; Tran, Lanh T. Local linear spatial regression. Ann. Statist., Tome 32 (2004) no. 1, pp.  2469-2500. http://gdmltest.u-ga.fr/item/1107794876/