Main Article Content
Abstract
In this paper we propose Efron residual based-bootstrap approximation methods in asymptotic model-checks for homoschedastic spatial linear regression models. It is shown that under some regularity conditions given to the known regression functions the bootstrap version of the sequence of least squares residual partial sums processes converges in distribution to a centred Gaussian process having sample paths in the space of continuous functions on 1,0 1,0 :I . Thus, Efron residual based-bootstrap is a consistent approximation in the usual sense. The finite sample performance of the bootstrap level Kolmogorov-Smirnov (KS) type test is also investigated by means of Monte Carlo simulation.