Search found 13307 matches

by EViews Gareth
Thu May 20, 2010 7:52 am
Forum: Data Manipulation
Topic: Extending a series
Replies: 2
Views: 3109

Re: Extending a series

Code: Select all

smpl 2008q4 2009q2
scalar growth = @mean(dlog(y))
smpl 2009q3 2009q4
y = (1+growth)*y(-1)


Something like that should do it.
by EViews Gareth
Thu May 20, 2010 6:48 am
Forum: Econometric Discussions
Topic: regression with population
Replies: 3
Views: 4234

regression with population

I'm not sure I understand what your question is.
by EViews Gareth
Wed May 19, 2010 10:19 pm
Forum: Installation and Registration
Topic: EViews 7 on Linux?
Replies: 8
Views: 43862

Re: EViews 7 on Linux?

EViews cannot collect valid machine ID information under Wine, which is why installation will not work. We've found no way around this.
by EViews Gareth
Wed May 19, 2010 7:55 am
Forum: Estimation
Topic: Chow Breakpoint Test - N/A Values
Replies: 1
Views: 2660

Re: Chow Breakpoint Test - N/A Values

Any observations with NAs would have been dropped from the original equation, and thus will not be part of the Chow test (or any other post-estimation test).
by EViews Gareth
Wed May 19, 2010 7:16 am
Forum: Add-in Support
Topic: EqTabs (Equation summaries)
Replies: 38
Views: 79138

EqTabs (Equation summaries)

It doesn't re-estimate the regressions. Could you post an example workfile where EViews crashes?
by EViews Gareth
Tue May 18, 2010 10:39 am
Forum: Programming
Topic: Reverse @elem(series, date)
Replies: 6
Views: 5381

Re: Reverse @elem(series, date)

The easiest way to do that is still with samples: smpl 2010m1 2010m1 y = y_sa(2) smpl @all or (not sure if this works, but should): smpl 2010m1 2010m1 y = @elem(y_sa,"2010m03") smpl @all If that doesn't work, you can break it up into two steps: smpl 2010m1 2010m1 !val = @elem(y_sa,"20...
by EViews Gareth
Tue May 18, 2010 8:19 am
Forum: Programming
Topic: Reverse @elem(series, date)
Replies: 6
Views: 5381

Re: Reverse @elem(series, date)

I'm still not clear on why my above solution doesn't do what you want.
by EViews Gareth
Tue May 18, 2010 8:18 am
Forum: Estimation
Topic: Stepwise Least Squares ->Lag specification
Replies: 6
Views: 6588

Re: Stepwise Least Squares ->Lag specification

Hard to say without having your workfile.
by EViews Gareth
Tue May 18, 2010 8:17 am
Forum: Estimation
Topic: near singular matrix error
Replies: 1
Views: 2999

Re: near singular matrix error

You've created a perfect collinearity problem. Total crime is equal to the sum of the other crimes.
by EViews Gareth
Mon May 17, 2010 2:30 pm
Forum: Programming
Topic: Reverse @elem(series, date)
Replies: 6
Views: 5381

Re: Reverse @elem(series, date)

Maybe I'm missing something, but don't you just mean:

Code: Select all

smpl 2010m5 2010m5
y = ysa
smpl @all
by EViews Gareth
Mon May 17, 2010 1:17 pm
Forum: Programming
Topic: Current Folder (EV6)
Replies: 4
Views: 4276

Re: Current Folder (EV6)

It is a little more complicated than that in general, but in your case, you'll probably find that @runpath works for you:

Code: Select all

%currpath = @runpath


If you move up to EViews 7, it is a lot more complicated, and you should consult our Whitepaper on paths:
http://eviews.com/download/download.html#docs
by EViews Gareth
Mon May 17, 2010 11:38 am
Forum: Suggestions and Requests
Topic: A few things about Database
Replies: 5
Views: 5839

Re: A few things about Database

If you're using the copy command, you can explicitly state the name of the object you want to create (i.e. rename it).

Code: Select all

copy fred::usa04405  localdb::pmi_chicago
by EViews Gareth
Mon May 17, 2010 11:36 am
Forum: Programming
Topic: Current Folder (EV6)
Replies: 4
Views: 4276

Re: Current Folder (EV6)

Define what you mean by "where the code is saved"
by EViews Gareth
Mon May 17, 2010 11:35 am
Forum: Data Manipulation
Topic: Exporting table to Excel
Replies: 10
Views: 20364

Re: Exporting table to Excel

No change.
by EViews Gareth
Sun May 16, 2010 9:47 pm
Forum: Estimation
Topic: Computing a Wald Test in a VAR
Replies: 22
Views: 38130

Re: Computing a Wald Test in a VAR

Yes it does work in EViews 6. You have to estimate the system first though.

Go to advanced search