The Stochastic Permanent Break Model and the Fractional Integration Hypothesis

In this article we show via simulations how the stochastic permanent break (STOPBREAK) model proposed by Engle and Smith (1999) is related with the fractionally integrated hypotheses. This connection was established by Diebold and Inoue (2001), showing, theoretically and analytically that stochastic regime switching is easily confused with long memory. In this paper, we use a version of the tests of Robinson (1994) for testing I(d) statistical models in the context of stochastic permanent break models, and give further evidence that both types of processes are easily confused.