Page 3 of 3
Re: Changing variable names in several equations
Posted: Thu Aug 02, 2018 11:30 am
by EViews Matt
Assuming for the moment that the observation you're dealing with is in program variable !j, you should be able to use something like:
Code: Select all
if @abs(res{%eq}(!j)) > {%eq}.@se then
%estcmd = {%eq}.@command + "@event(""" + @otod(!j) +""")"
Re: Changing variable names in several equations
Posted: Thu Aug 02, 2018 12:26 pm
by statsforecast
Thank you Matt! I imagined that I would have to loop through the observations in each residual series. Any idea of how to do this?
Re: Changing variable names in several equations
Posted: Thu Aug 02, 2018 1:28 pm
by EViews Matt
The following should go through every observation in the current sample. !j is the workfile observation as previously mentioned.
Code: Select all
for !k = 1 to @obssmpl
!j = @dtoo(@otods(!k))
...
next
Re: Changing variable names in several equations
Posted: Tue Aug 07, 2018 8:48 am
by statsforecast
Thank you Matt!
Re: Changing variable names in several equations
Posted: Wed Aug 15, 2018 12:51 pm
by statsforecast
I am trying to change the variables names in a group of series when writing to excel. I usually have something like this:
write(e,noid) "filepath" DATE group01
I would like to perform something like @mid(group01.@members, 3,8) to each series name in group01 when outputting to excel. What I am trying to do basically is to fetch the original series names which were numerical (8 digits), and strip them from the alpha characters that Eviews adds to make the series names valid and the character f at the end of the forecast series.
I would appreciate any ideas.
Thank you.