about estimate equation and BDS test

For technical questions regarding estimation of single equations, systems, VARs, Factor analysis and State Space Models in EViews. General econometric questions and advice should go in the Econometric Discussions forum.

Moderators: EViews Gareth, EViews Moderator

tg128
Posts: 18
Joined: Mon Sep 22, 2008 1:10 pm

about estimate equation and BDS test

Postby tg128 » Tue Sep 23, 2008 8:45 am

Just wondering if there is anyway I can estimate more than one equation in the estimate equation window please??

another question, when use BDS test, there are both BDS statistic and z-statistic, on what ground I should reject the null hypothesis of the BDS test?? what is v-statistic??

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13585
Joined: Tue Sep 16, 2008 5:38 pm

Postby EViews Gareth » Tue Sep 23, 2008 8:48 am

You cannot estimate more than one equation in an equation object. You'll have to use a System.

tg128
Posts: 18
Joined: Mon Sep 22, 2008 1:10 pm

Postby tg128 » Tue Sep 23, 2008 8:56 am

You cannot estimate more than one equation in an equation object. You'll have to use a System.
right, what if I want to compute information criteria to select ARMA model, how can I use system to do that please?

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13585
Joined: Tue Sep 16, 2008 5:38 pm

Postby EViews Gareth » Tue Sep 23, 2008 9:05 am

You'll have to compute the ICs yourself, using the determinant residual covariance to compute the likelihood.

EViews Chris
EViews Developer
Posts: 161
Joined: Wed Sep 17, 2008 10:39 am

Re: about estimate equation and BDS test

Postby EViews Chris » Thu Sep 25, 2008 5:46 pm

another question, when use BDS test, there are both BDS statistic and z-statistic, on what ground I should reject the null hypothesis of the BDS test?? what is v-statistic??
The z-statistic is the BDS-statistic divided by the standard error. It is the final result that is used for the hypothesis test (because it is asymptotically standard normal).

EViews reports the probability that the given z-statistic would be observed from iid data in the Prob. column of the output, so you would reject the hypothesis that the series is iid if the reported probability is small.

The V-statistics are intermediate results used in calculating some of the other numbers. You can ignore them if you're only interested in the final results.


Return to “Estimation”

Who is online

Users browsing this forum: No registered users and 2 guests