TY - JOUR
T1 - Efficient tests for general persistent time variation in regression coefficients
AU - Elliott, Graham
AU - Müller, Ulrich K.
N1 - Funding Information:
where the weak convergence follows from the uniform convergence of T t=1−1 ∑[sT] Qt∗ Qt∗′ →p sIk+d , the consistency of VˆX, the CMT and the FCLT for mixing series as in the proof of Lemma 3. Proceeding as in the proof of Lemma 2 now yields the result. ‖ Acknowledgements. Graham Elliott is grateful to the NSF for financial assistance under grant SES 0111238, and Ulrich Müller gratefully acknowledges the financial support of the Swiss National Science Foundation. The authors thank two anonymous referees, the editor, Bernard Salanié, and Mark Watson, Giorgio Primiceri, Allan Timmermann, Paolo Giordani, Piotr Eliasz, and seminar participants at EUI, Berkeley, UCSD, Princeton, Boston University, LSE, Aarhus, Yale, Harvard, and at the ReStud Tour 2003 for helpful comments.
PY - 2006/10
Y1 - 2006/10
N2 - There are a large number of tests for instability or breaks in coefficients in regression models designed for different possible departures from the stable model. We make two contributions to this literature. First, we consider a large class of persistent breaking processes that lead to asymptotically equivalent efficient tests. Our class allows for many or relatively few breaks, clustered breaks, regularly occurring breaks, or smooth transitions to changes in the regression coefficients. Thus, asymptotically nothing is gained by knowing the exact breaking process of the class. Second, we provide a test 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 good power and size properties even in heteroscedastic models.
AB - There are a large number of tests for instability or breaks in coefficients in regression models designed for different possible departures from the stable model. We make two contributions to this literature. First, we consider a large class of persistent breaking processes that lead to asymptotically equivalent efficient tests. Our class allows for many or relatively few breaks, clustered breaks, regularly occurring breaks, or smooth transitions to changes in the regression coefficients. Thus, asymptotically nothing is gained by knowing the exact breaking process of the class. Second, we provide a test 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 good power and size properties even in heteroscedastic models.
UR - http://www.scopus.com/inward/record.url?scp=33749072972&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=33749072972&partnerID=8YFLogxK
U2 - 10.1111/j.1467-937X.2006.00402.x
DO - 10.1111/j.1467-937X.2006.00402.x
M3 - Article
AN - SCOPUS:33749072972
VL - 73
SP - 907
EP - 940
JO - Review of Economic Studies
JF - Review of Economic Studies
SN - 0034-6527
IS - 4
ER -