On the sample variance of explosive random coefficient autoregressive processes

This work investigates the behavior of the sample variance of an explosive random coefficient autoregressive process yt=(a+ut)yt−1+εtyt=(a+ut)yt−1+εt. It is shown that the simulated sample variance has a distribution when a2<1a2<1 and a2+σu2=1. Moreover, the variance of ytyt when a=−1a=−1 is found to be three times larger as compared with the case where a=1a=1.