The Statistical Inferences On Change Point Of Panel Data

The change point problem has been a hot topic in finance and econometrics since the 1970 s.The application of panel data in finance is very extensive.The application of change point research method to panel data can be used as a means of depth study of change point theory.Therefore,on the basis of the existing research results of change points,this paper generalizes the change points theory in time series data to panel data,and mainly studies the estimation and testing of mean and variance change points theoretically.The specific research contents are as follows.Firstly,The mean change point of panel data is studied by Cumulative Sum Method.The estimator of change point is given.The strong consistency of the estimator of the change point is proved and the convergence rates of the change point estimator is given.The problem of mean change point test is studied,The ratio test statistic based on cumulative Sum functional is constructed.Under the null hypothesis the asymptotically limiting distribution of the statistic is obtained.Secondly,The problem of variance change point of panel data is studied.The estimators of the variance change point is proposed.The consistency of the estimator of the change point is proved.The problem of variances change point test is studied,the ratio test statistic is given,the limiting distribution of the test under the null hypothesis is derived.Finally,The work done in this paper is summarized and the contents that need to be further studied are put forward.