AI Summary & Key Takeaways
AI InsightA simulation as well as an empirical analysis is also carried out to get more generalized view on the model under study.
📌 3 Key Takeaways
Abstract
The idea about structural break in unit root hypothesis under time series model had received great amount of attention over many last decades. The importance of structural break in the mean had been comprehensively studied by Perron [15], Perron and Vogelsang [17], Zivot and Andrews [25] etc. This had also studied in considering of break in variance by Kim et al. [9], Cook [6], Kumar et al. [11] etc. There is sufficient contribution regarding break in mean and variance individually but both are equally important and this was little explored by Bai [1] for panel data and Meligkotsidou et al. [14] for univariate time series. In present paper, we are extending this on panel dataAR(1) time series model under Bayesian framework. Posterior odds ratio has been derived for various models with and without break in mean, variance and both in consideration of unit root hypothesis. A simulation as well as an empirical analysis is also carried out to get more generalized view on the model under study.