Title: Detection of changes in panel data models with stationary regressors
Authors: Charl Pretorius - North-West University (South Africa) [presenting]
Marie Huskova - Charles University (Czech Republic)
Abstract: A panel regression model with cross-sectional dimension $N$ is considered. The aim is to test, based on $T$ observations, whether the $N$ intercepts in the model remain unchanged throughout the observation period. The test procedure involves the use of a CUSUM-type statistic derived from a pseudo-likelihood argument. We present asymptotic results of the test statistic in the case where both $N$ and $T$ are allowed to become large. The asymptotic results are valid under strong mixing and stationarity assumptions on the error and regressor sequences. Monte Carlo results will be presented that indicate that the tests work in the case of small to moderate sample sizes. The talk ends with an illustrative application of the procedure to financial data.