Page 1 of 1

Tse, JOE 2000 test for CC

Posted: Thu Jan 22, 2015 10:20 am
by TomDoan
This demonstrates the Tse LM test for constant correlation, replicating the results results in Tse, Y. K.(2000), "A Test for Constant Correlations in a Multivariate GARCH Model", Journal of Econometrics 98, 107-127. This starts with an estimate of a CC model and does an LM test for whether the covariances seem to also depend upon lagged outer products of the residuals.

This uses the @TSECCTEST procedure. Note that the GARCH model cannot use off-diagonal covariances in either the mean or variance models. The use of variances is OK, so you can have an "M" model which uses variances, but not the covariances.
tse.rpf
Program file
(1.18 KiB) Downloaded 1049 times
exratew.dat
Data file
(19.03 KiB) Downloaded 1092 times

Re: Tse, JOE 2000 test for CC

Posted: Fri Jan 18, 2019 12:45 pm
by abi
Dear Tom,

What is null in Tse test?

Re: Tse, JOE 2000 test for CC

Posted: Fri Jan 18, 2019 1:01 pm
by TomDoan
CC.

Re: Tse, JOE 2000 test for CC

Posted: Fri Jan 18, 2019 1:12 pm
by abi
TomDoan wrote:CC.
Thank's Tom,

If the null is rejected then it means i should use DCC rather than CC or only i can not use CC?

Re: Tse, JOE 2000 test for CC

Posted: Fri Jan 18, 2019 2:18 pm
by TomDoan
No. It only means that CC doesn't seem to be adequate. Whether any specific alternative is better isn't clear.