Search found 11902 matches

by EViews Gareth
Thu Nov 29, 2018 9:28 am
Forum: Data Manipulation
Topic: reshape panel data to pooled data
Replies: 2
Views: 293

Re: reshape panel data to pooled data

The issue is that you have a series IFI, and IFI1, IFI2, IFI3, and that your cross-sections are labelled 1,2,3,... When you do a standard unstack, the series IFI will be unstacked to IFI1, IF2, IF3, ... but they already exist as other series. Renaminng IFI1,2,3 in the source page should fix the issue.
by EViews Gareth
Wed Nov 28, 2018 10:43 am
Forum: Data Manipulation
Topic: import excel files with wildcard in names
Replies: 2
Views: 303

Re: import excel files with wildcard in names

Use the @wdir function
by EViews Gareth
Sat Nov 24, 2018 3:22 pm
Forum: Programming
Topic: Data Series Label
Replies: 1
Views: 1523

Re: Data Series Label

It isn't built in, but you can hack it by creating a series that is all NA apart from the final value on the graph. Then graph that series at the same time as the original series, and set the line style to "obs label". Then remove its label entry.
by EViews Gareth
Fri Nov 23, 2018 2:24 pm
Forum: Programming
Topic: Graph Label Command
Replies: 2
Views: 93

Re: Graph Label Command

Something like:

Code: Select all

%text = "Data as of: " + @strnow("YYYY-Mon-DD")
g.addtext(t) {%text}
by EViews Gareth
Fri Nov 23, 2018 2:22 pm
Forum: Programming
Topic: Workfile Range Command
Replies: 2
Views: 362

Re: Workfile Range Command

You were close.

Code: Select all

wfcreate FX_D d(1,5) 01/01/1980 @now
by EViews Gareth
Thu Nov 22, 2018 11:20 pm
Forum: General Information and Tips and Tricks
Topic: Stars on coefficients' significance
Replies: 1
Views: 3675

Re: Stars on coefficients' significance

Take a look at the signifcoefs add-in. It does something close to what you want (shading significance rather than using *), but you should get an idea of how it works and modify it easily enough.
by EViews Gareth
Thu Nov 22, 2018 6:11 pm
Forum: Programming
Topic: ERRORHELP
Replies: 42
Views: 14483

Re: ERRORHELP

It is Thanksgiving. Don’t expect any responses for a couple of weeks from EViews staff.
by EViews Gareth
Wed Nov 21, 2018 10:02 pm
Forum: General Information and Tips and Tricks
Topic: variable using previous year data
Replies: 3
Views: 756

Re: variable using previous year data

Create the workfile with however much data you have. EViews will figure out how much it can use during estimation itself.
by EViews Gareth
Wed Nov 21, 2018 7:58 pm
Forum: Programming
Topic: FX Cross Rates Program
Replies: 4
Views: 447

Re: FX Cross Rates Program

Something like:

Code: Select all

for %i eur jpy chf gbp sek nok aud nzd cad
for %j eur jpy chf gbp sek nok aud nzd cad
if %j<>%i then
frml {%i}{%j} = {%i}/{%j}
endif
next
next


That will give you the pairs in both directions.
by EViews Gareth
Wed Nov 21, 2018 7:27 pm
Forum: General Information and Tips and Tricks
Topic: variable using previous year data
Replies: 3
Views: 756

Re: variable using previous year data

Assuming you have annual data, enter the variable as consumption(-1)
by EViews Gareth
Wed Nov 21, 2018 12:45 pm
Forum: Programming
Topic: FX Cross Rates Program
Replies: 4
Views: 447

Re: FX Cross Rates Program

What does "calculate each rate versus each other" mean?
by EViews Gareth
Thu Nov 15, 2018 9:00 am
Forum: Programming
Topic: Series with different range
Replies: 2
Views: 595

Re: Series with different range

Easiest way, if the series don't have interior missings, is to just do:

Code: Select all

smpl if s1<>na
smpl if s2<>na

(or in a loop).

If there are interior missings, then something like:

Code: Select all

%start  = s1.@first
%end = s1.@last
smpl {%start} {%end}
by EViews Gareth
Thu Nov 15, 2018 8:33 am
Forum: Data Manipulation
Topic: How is/How to generate constant (other than 0's)?
Replies: 1
Views: 596

Re: How is/How to generate constant (other than 0's)?

The "C" object in the workfile is the coefficient vector. It contains the results of the last estimation performed. In a new workfile, if you haven't run any regressions yet, it will be full of zeroes. As soon as you estimate something, it will become populated. The values in the "C&q...
by EViews Gareth
Wed Nov 14, 2018 10:15 am
Forum: Data Manipulation
Topic: NAs in sample
Replies: 8
Views: 318

Re: NAs in sample

No.

I still don't see what you're trying to do. If you just want a count,why do you need to see the series?

Go to advanced search