A1005
Title: Online break-detection for panel regression models
Authors: Marie Huskova - Charles University (Czech Republic)
Charl Pretorius - North-West University (South Africa) [presenting]
Abstract: The aim is to propose a CUSUM-type test procedure for monitoring sequential panel data to check for any changes in panel means. The limit behavior of the procedure under suitable weak dependence conditions on the errors and regressors is presented. The results are extended to the case where there exists strong cross-sectional dependence modeled by means of the common factor model. A self-normalized version of the procedure is also proposed, which is convenient from a practical perspective particularly to avoid the estimation of long-run variances. The theoretical results are supported by a simulation study that indicates that the procedure works well in the case of small to moderate sample sizes. An illustrative application is presented.
