Search found 429 matches

by EViews Matt
Mon Jun 22, 2020 1:59 pm
Forum: Programming
Topic: Rolling regressions with conditions
Replies: 5
Views: 310

Re: Rolling regressions with conditions

Hello,

Your extra requirements would be relatively minor changes to a rolling regression program, (1) a new outer loop to go through each of your 100 stocks, and (2) an if statement to check whether you should skip the current regression if there are fewer than 24 non-NA observations available.
by EViews Matt
Tue Jun 09, 2020 9:23 am
Forum: Programming
Topic: Hessian is singular at iteration 1
Replies: 2
Views: 213

Re: Hessian is singular at iteration 1

Hello,

That error is generated by one of the FRB/US model program subroutines (in mce_solve_library.prg) rather than by EViews. I haven't been able to recreate this issue running the stock ocpolicy.prg with the unmodified database.
by EViews Matt
Sun Jun 07, 2020 7:56 am
Forum: Estimation
Topic: SVAR with scenarios
Replies: 7
Views: 3388

Re: SVAR with scenarios

Hello,

Not automatically, no. You'd have to perform the calculations for such analyses yourself.
by EViews Matt
Sun Jun 07, 2020 7:51 am
Forum: Programming
Topic: Adding a new variable in the FRB/US data in EViews 11
Replies: 3
Views: 329

Re: Adding a new variable in the FRB/US data in EViews 11

Hello, Regarding future data, if you can't find someone else's forecast, you can always make your own. Many of the series bundled with the FRB/US model simply use the last known value for all future values. Regarding the %sufsim = "_1" line, if you look at how that program variable is used...
by EViews Matt
Thu Jun 04, 2020 6:54 pm
Forum: Bug Reports
Topic: Error in using FRB/US model in Eviews 11
Replies: 6
Views: 467

Re: Error in using FRB/US model in Eviews 11

Hello, It just occurred to me that this may be a simple working directory issue, as the FRB/US programs are written under the assumption that the working directory is that package's "programs" directory. If that's the problem, just set the working directory accordingly (double-click the &q...
by EViews Matt
Fri May 22, 2020 4:12 pm
Forum: Data Manipulation
Topic: problems with @during with hourly frequency
Replies: 6
Views: 446

Re: problems with @during with hourly frequency

There does seem to be an issue with BREAKLS in an intraday workfile. It doesn't appear to be related to @during, but I'm still investigating. I don't have a workaround for you, other than perhaps to use an undated workfile for the time being.
by EViews Matt
Fri May 22, 2020 2:14 pm
Forum: Data Manipulation
Topic: problems with @during with hourly frequency
Replies: 6
Views: 446

Re: problems with @during with hourly frequency

Hello,

It appears that @during doesn't accept intraday date notation. For the time being, you can work around this by using an @after and @before clause instead, e.g.

Code: Select all

series ind = @after("04/04/2020 01:00") and @before("04/06/2020 02:00")
by EViews Matt
Thu May 14, 2020 9:33 am
Forum: Models
Topic: How to simulate an articial increase in the 5, 10, and 30 -year interest rates in the FRB/US Model
Replies: 2
Views: 747

Re: How to simulate an articial increase in the 5, 10, and 30 -year interest rates in the FRB/US Model

Hello, I assume you're referring to model variables RG5, RG10, and RG30. A simple way to explore this type of alteration to the model is to exclude and override any or all of those variables. Excluding a variable causes it to be treated as exogenous, and overriding a variable lets you provide your o...
by EViews Matt
Tue May 12, 2020 11:26 am
Forum: Data Manipulation
Topic: Row index of the n-highest value
Replies: 1
Views: 215

Re: Row index of the n-highest value

Hello, You can construct a matrix where the n-th column holds the indices of the n-th largest value in each row. You could then utilize the column of interest. For example, for a group named "g", matrix tmp1 = @transpose(@colcumsum(@ones(@columns(g), @rows(g)))) matrix tmp2 = @rowranks(@co...
by EViews Matt
Tue May 05, 2020 1:38 pm
Forum: Bug Reports
Topic: @addquotes gives error if applied to alpha series
Replies: 1
Views: 822

Re: @addquotes gives error if applied to alpha series

Hello,

This will be fixed in the next patch.
by EViews Matt
Wed Apr 22, 2020 11:39 am
Forum: Estimation
Topic: SVAR model
Replies: 1
Views: 399

Re: SVAR model

Hello, (1) Yes in the sense that they both produce a recursive factorization. If you're asking whether using the Cholesky decomposition versus estimating an A-B SVAR decomposition will produce exactly the same results, that's a more subtle issue. Theoretically, the two techniques can match, and like...
by EViews Matt
Fri Apr 10, 2020 7:07 am
Forum: Programming
Topic: setmaxerrs - syntax error in control statement
Replies: 6
Views: 674

Re: setmaxerrs - syntax error in control statement

Ah, I see the problem scenario now. When %Variables is empty because the assignment to it has failed, then the for statement effectively becomes just "for %Var", which is an ill-formed expression and a critical error. Since you cannot suppress this error, your best bet is to alter the prog...
by EViews Matt
Thu Apr 09, 2020 12:52 pm
Forum: Programming
Topic: setmaxerrs - syntax error in control statement
Replies: 6
Views: 674

Re: setmaxerrs - syntax error in control statement

While there's no way to suppress critical errors, neither of the errors you've mentioned are in the critical category. Is it possible setmaxerrs simply isn't set large enough?
by EViews Matt
Thu Apr 09, 2020 6:47 am
Forum: Programming
Topic: setmaxerrs - syntax error in control statement
Replies: 6
Views: 674

Re: setmaxerrs - syntax error in control statement

Hello,

There are about two dozen "critical" errors that halt an EViews program even when setmaxerrs is being used. Most of those errors relate to syntax errors, which cannot be safely ignored and should be fixed in the program. What is the exact error you're receiving?

Go to advanced search