## Search found 327 matches

Wed Jan 16, 2019 12:57 pm
Forum: Econometric Discussions
Topic: residual matrix
Replies: 3
Views: 123

### Re: residual matrix

Begin by reviewing the SVAR section of the EViews documentation. You're going to have to choose what SVAR model you want. Once you've successfully estimated the SVAR model, you can then generate the structural residual series and finally calculate the covariance matrix for those series.
Wed Jan 16, 2019 11:04 am
Forum: Econometric Discussions
Topic: residual matrix
Replies: 3
Views: 123

### Re: residual matrix

Hello, The residuals will never be perfectly orthogonal. If you're still working with the synthetic w y model you've posted in other forums, you can even see that your randomly generated innovations e and u are not perfectly orthogonal (just execute "cov u e"). You can certainly apply an S...
Wed Jan 16, 2019 10:17 am
Forum: Programming
Topic: Programming the renaming of multiple variables
Replies: 1
Views: 45

### Re: Programming the renaming of multiple variables

Hello,

You can generate lists of EViews objects using @wlookup and the use various string manipulation functions to build your rename statements. It all depends on how complicated the renaming patterns are.
Tue Jan 15, 2019 3:00 pm
Forum: Programming
Topic: Matrix of autocovariance of a var
Replies: 1
Views: 91

### Re: Matrix of autocovariance of a var

Hello, EViews is more than capable of calculating the autocovariances and crosscovariances for you, so there's no need to loop through the data yourself. Here's a quick way to calculate what you want, group g w y w(-1) y(-1) g.cov(out=tmp) matrix ac0 = @subextract(tmpcov, 1, 1, 2, 2) matrix ac1 = @s...
Tue Jan 15, 2019 10:38 am
Forum: Programming
Topic: Represent a var (2) as var(1
Replies: 1
Views: 63

### Re: Represent a var (2) as var(1

Hello,

You can create a VAR of any order you wish. The default lag interval is "1 2", creating a VAR with lags for t-1 and t-2, but you can change the interval to "1 1" to create a VAR with lags for only t-1.
Thu Jan 10, 2019 11:15 am
Forum: Estimation
Topic: Triangular Factorization (non-orthonormal covariance matrix)
Replies: 3
Views: 112

### Re: Triangular Factorization (non-orthonormal covariance matrix)

Mathematically, it's not clear to me that there's any practical difference between the Cholesky and triangular factorization. Using what EViews calls an A-B SVAR model, i.e. A * y = <regressors> + B * u, you can certainly "pretend" that the structural innovations have diagonal covariance b...
Tue Jan 08, 2019 10:54 am
Forum: Estimation
Topic: Convergence achieved after 0 iterations
Replies: 1
Views: 63

### Re: Convergence achieved after 0 iterations

Hello,

That is suspicious. Either your starting values lucked upon a likelihood maximum, or there's an optimization failure that's not being reported correctly. Could you provide a workfile with your data and estimation?
Tue Jan 08, 2019 10:49 am
Forum: Programming
Topic: perform a regression for each crossid in a panel workfile and store the slopes
Replies: 2
Views: 87

### Re: perform a regression for each crossid in a panel workfile and store the slopes

Hello,

You just need to create the auxiliary storage for the betas and then copy them out of the equation objects one at a time.

Code: Select all

`vector(1200) betasfor !i=1 to 1200   smpl if @crossid=!i   equation linreg{!i}.LS Y C X1 X2   betas(!i) = linreg{!i}.@coefs(2)next`
Fri Jan 04, 2019 2:53 pm
Forum: Estimation
Topic: Triangular Factorization (non-orthonormal covariance matrix)
Replies: 3
Views: 112

### Re: Triangular Factorization (non-orthonormal covariance matrix)

Hello, What you suggest could cause a change in the magnitude of the impulse responses, but not their overall shape. EViews' assumption that the structural innovations are orthonormal is reflected in the structural impulse response generation algorithm that uses unit (magnitude one) structural innov...
Thu Jan 03, 2019 10:26 am
Forum: Estimation
Topic: Hours of the weekday dummy
Replies: 2
Views: 97

### Re: Hours of the weekday dummy

Trying to combine a subset of those dummy series is more complicated than just creating the dummies you want from scratch... group dummies string dayNames = "M Tu W Th F" for !day = 1 to 5 for !hour = 0 to 23 %name = "D" + @word(dayNames, !day) + "_" + @str(!hour, "...
Thu Dec 20, 2018 7:05 pm
Forum: Programming
Topic: ERRORHELP
Replies: 46
Views: 15454

### Re: ERRORHELP

Okay, that's what my suggested change to !fake_aic should accomplish.
Thu Dec 20, 2018 10:34 am
Forum: Programming
Topic: ERRORHELP
Replies: 46
Views: 15454

### Re: ERRORHELP

If I understand you correctly, you now want only two cases: 0 or 1 p-values < 0.1, and 2 p-values < 0.1, correct? A small change to the previous !fake_aic calculation can accomplish that. !fake_aic = eq{!i}.@aic - 1000 * (eq{!i}.@pval(2) < !pval_C and eq{!i}.@pval(3) < !pval_D)
Tue Dec 18, 2018 5:02 pm
Forum: Programming
Topic: avoid "divide by zero" prg stop
Replies: 1
Views: 81

### Re: avoid "divide by zero" prg stop

Hello, You can increase the number of errors allowed before EViews terminates the program (the default is one). The "Run Program" dialog has a field for "Maximum errors before halting:", which you can simply change to a large value. Alternatively, within a program you can adjust ...
Fri Dec 14, 2018 3:54 pm
Forum: Programming
Topic: regression loop doesn't work
Replies: 2
Views: 145

### Re: regression loop doesn't work

Hello,

If all you need is to run those five estimations, you were making the program much more complicated than it needed to be:

Code: Select all

`for %n G_MA3 I_MA3 M_MA3 MS_MA3 X_MA3   equation eq_{%n}.ls @pcy({%n}) c @pcy({%n}_TC)next`
Mon Dec 10, 2018 10:20 am
Forum: Programming
Topic: ERRORHELP
Replies: 46
Views: 15454

### Re: ERRORHELP

My point is that the estimates you're evaluating to find the best combination of lags (the "equation eq{!i}.ls" line) and the estimate you're using to record the results for the best combination (the "show eq{!i}.ls" line) do not match. In the former you use the "to" la...