演讲人：Bin Chen（University of Pittsburgh）
时间：2008年5月8日，周四, 15:30-17:00
地点：上海交通大学安泰楼102
演讲内容梗概：Checking parameter stability of economic models is a long-standing problem in time series econometrics. A classical econometric procedure is Chow’s (1960) test, which checks for the existence of a structural change on a known date. Various extensions have been made to test multiple changes with known or unknown change points. However, almost all existing structural change tests in econometrics are deigned to detect abrupt breaks. Little attention has been paid to smooth structural changes, which may be more realistic in economics. This paper proposes two consistent tests for smooth structural changes as well as abrupt structural breaks with known or unknown change points. The idea is to estimate the smooth time-varying parameters bylocal smoothing and compare them with the OLS parameter estimator. The first test compares the sums of squared residuals of the restricted constant parameter model and the unrestricted nonparametric time-varying parameter model, in a spirit similar to Chow’s (1960) F test. The second test compares the fitted values of the restricted and unrestricted models, which can be viewed as a generalization of Hausman’s (1978) test. Both tests have a convenient asymptotic N(0,1) distribution and do not require any prior information about the alternatives. Interestingly, unlike Chow’s (1960) test, the generalized Chow test is no longer optimal; it is asymptotically less powerful than the generalized Hausman test. A simulation study highlights the merits of the proposed tests in comparison with a variety of popular tests for structural changes.
