演講者簡介 : Professor Rachinger received his Ph.D. in Economics from Universidad Carlos III at 2012. He is currently an Assistant Professor at Department of Economics at University of Vienna. His research fields are Econometrics, Time Series Analysis, Macroeconometrics.
演講摘要 : We analyze least squares (LS) estimation of breaks in long memory time series. We show that the estimator of the break fraction is consistent and converges at rate T when there is a break in the level, in the memory or in both parameters. We also derive the asymptotic distribution of the break fraction under shrinking break magnitudes. Further, we analyze tests for the number of breaks. When testing for breaks in the memory, the asymptotic results correspond to standard ones in the literature. When testing for breaks in the level and when testing for breaks in both parameters, the results differ in terms of the asymptotic distribution of the test statistic. In this case, the LS-procedure loses its asymptotic pivotality. We further propose a method in order to distinguish between long memory, breaks in the memory and breaks in the level. Such a distinction is difficult but is important for reasons such as shock identification, forecasting and detection of spurious fractional cointegration. In a simulation exercise, we find that the tests based on asymptotic critical values are oversized in finite samples. Therefore, we suggest using the bootstrap, for which we derive validity and consistency, and we confirm its better size properties. Finally, we use the method to test for breaks in the U.S. inflation rate.
