Template-Type: ReDIF-Paper 1.0

Author-Name: Zhongjun Qu
Author-X-Name-First: Zhongjun 
Author-X-Name-Last: qu
Author-Email: qu@bu.edu.
Author-Workplace-Name: Department of Economics, Boston University
.
Author-Name: Pierre Perron
Author-X-Name-First: Pierre Perron
 Author-X-Name-Last:Perron
Author-Email: Perron@bu.edu
Author-Workplace-Name: Department of Economics, Boston University



Title: Estimating and testing structural changes in
multivariate regressions



Abstract: This paper considers issues related to estimation, inference and computation with
multiple structural changes occurring at unknown dates in a system of equations.
Changes can occur in the regression coefficients and/or the covariance matrix of the
errors. We also allow arbitrary restrictions on these parameters, which permits the
analysis of partial structural change models, common breaks occurring in all equations,
breaks occurring in a subset of equations, etc. The method of estimation is quasi
maximum likelihood based on Normal errors. The limiting distributions are obtained
under more general assumptions than previous studies. Of special interest is the fact
that substantial efficiency gains can be obtained by casting a regression affected by
changes in a system even if the other equations are not affected by breaks, provided
there is non-zero correlation between the errors. For testing, we propose likelihood
ratio type statistics to test the null hypothesis of no structural change and to select the
number of changes. Structural change tests with restrictions on the parameters can
be constructed to achieve higher power when prior information is present. We propose
an algorithm for an efficient procedure to construct the estimates and test statistics.
We also introduce a novel locally ordered breaks model, which allows the breaks in
different equations to be related yet not occurring at the same dates.
.


Length: 75 pages
Creation-Date: 2005-03
Revision-Date: 
Publication-Status:
File-URL:http://www.bu.edu/econ/workingpapers/papers/Pierre%20Perron/wp2005/qu-perron.pdf
File-Format: Application/pdf
File-Function: 
Number: WP2005-012
Classification-JEL: 
Keywords: change-point, segmented regressions, break dates, hypothesis testing,
model selection, system of regressions.
Handle: RePEc:bos:wpaper:WP2005-012