Optimally Testing General Breaking Processes in Linear Time Series Models

There are a large number of tests for instability or breaks in coefficients in regression models designed for different possible departures from a stable regression. We make two contributions to this literature. First, we provide conditions under which optimal tests are asymptotically equivalent. Our conditions allow for models with many or relatively few breaks, clustered breaks, regularly occurring breaks or smooth transitions to changes in the regression coefficients. Thus we show nothing is gained asymptotically by knowing the exact breaking process. Second, we provide a statistic that is simple to compute, avoids any need for searching over high dimensions when there are many breaks, is valid for a wide range of data generating processes and has high power for many alternativ