Hi,
I am working with Eviews 7.2
I found one bug about the command "makeresid".
If I try to create the residuals series from the new constructed cointegration function with dynamic ols, the eviews would exit aotumatically.
equation eq1.cointreg(method=dols, lltype=none) log(variable 1) log(variable 2) ...
eq1.makeresid res
But, this command works for the other type of regression.
for example,
equation eq1.cointreg variable1 variable2
eq1.makeresid res
If I created one cointegration function with dols first, then save it, and then exit. Then, log in the eviews agian. The residuals still can be gained.
The problem for me is that now I have 14641 pair-wises for ECM analysis. I cannot save the residuals manually one by one.
My programme cannot run because of this bug (my programme can run for the other type of regression).
Could you fix it in your next patch or tell me how to deal with this problem with eviews 7.2
many thanks
Jianan Li
makeresid for dynamic ols doesn't work
Moderators: EViews Gareth, EViews Moderator
-
EViews Glenn
- EViews Developer
- Posts: 2682
- Joined: Wed Oct 15, 2008 9:17 am
Re: makeresid for dynamic ols doesn't work
We're looking into it...
-
EViews Glenn
- EViews Developer
- Posts: 2682
- Joined: Wed Oct 15, 2008 9:17 am
Re: makeresid for dynamic ols doesn't work
Forgot to post an update. Sorry.
The problem was related to the fact that the user's model was a static (no leads, lags, differences) DOLS specification (I like that, "static dynamic ols" :)). We weren't handling the zero augmenting terms case properly when trying to make the residuals.
A fix for this was posted on 7/26. Sorry for the inconvenience.
The problem was related to the fact that the user's model was a static (no leads, lags, differences) DOLS specification (I like that, "static dynamic ols" :)). We weren't handling the zero augmenting terms case properly when trying to make the residuals.
A fix for this was posted on 7/26. Sorry for the inconvenience.
Re: makeresid for dynamic ols doesn't work
many thanks for your help :D
Forgot to post an update. Sorry.
The problem was related to the fact that the user's model was a static (no leads, lags, differences) DOLS specification (I like that, "static dynamic ols" :)). We weren't handling the zero augmenting terms case properly when trying to make the residuals.
A fix for this was posted on 7/26. Sorry for the inconvenience.
Who is online
Users browsing this forum: No registered users and 2 guests
