Torben Klarl
Is Spatial Bootstrapping a Panacea for Valid
Inference?
Abstract:
Bootstrapping methods have so far been rarely used to evaluate spatial
data sets. Based on an extensive Monte Carlo study we find that also
for spatial, cross-sectional data, the wild bootstrap test proposed by
Davidson and Flachaire (2008) based on restricted residuals clearly
outperforms asymptotic as well as competing bootstrap tests, like the
pairs bootstrap.
JEL: C18, C21, R11
Paper:
Paper available as pdf-file.
Beitrag Nr. 322, Volkswirtschaftliche Diskussionsreihe, Institut
für
Volkswirtschaftslehre der Universität Augsburg
Contact:
Torben
Klarl, University of Augsburg, Department of Economics,
D-86135
Augsburg,
Germany, phone +49-821-598-4189, fax +49-821-598-4231
email: torben.alexander.klarl@wiwi.uni-augsburg.de
Bo., 27.05.2013