Professor Peter Michael Robinson is currently Tooke Professor of Economic Science and Statistics at University of London (London School of Economics), a Co-Editor of Journal of Econometrics and Journal of Time Series Analysis and also a Associate Editor of Annals of the Statistics. Professor Robinson has been elected Fellow of the Econometric Society, Fellow of the Institute of Mathematical Statistics, and Fellow of the British Academy. Professor Robinson’s research interests focus on Econometrics, Time Series Analysis, Nonparametric Inference, Semiparametric Inference and Spatial Econometrics. His research articles have been published in several prestigious journals, such as Econometrica, Annals of Statistics, Journal of the American Statistical Association and Journal of Econometrics.
講題摘要:
We consider testing the null hypothesis of no spatial autocorrelation against the alternative of first order spatial autoregression. A Wald test statistic has good first-order asymptotic properties, but these may not be relevant in small or moderate-sized samples, especially as (depending on properties of the spatial weight matrix) the usual parametric rate of convergence may not be attained. We thus develop tests with more accurate size properties, by means of Edgeworth expansions and the bootstrap. The finite-sample performance of the tests is examined in Monte Carlo simulations.