Search found 3406 matches

by startz
Fri Jan 18, 2019 1:47 pm
Forum: Estimation
Topic: State Space - generate plot of unknown parameters.
Replies: 1
Views: 99

Re: State Space - generate plot of unknown parameters.

c(4) is an estimated constant. It doesn't very over time.
by startz
Mon Jan 14, 2019 2:15 pm
Forum: Econometric Discussions
Topic: High P Values
Replies: 1
Views: 49

Re: High P Values

You only have 13 observations. Probably nothing is wrong.
by startz
Thu Jan 10, 2019 3:55 pm
Forum: Estimation
Topic: Different results for running state space models
Replies: 3
Views: 112

Re: Different results for running state space models

Starting values (in the C vector) matter.
by startz
Tue Jan 01, 2019 9:26 am
Forum: Data Manipulation
Topic: import panel data
Replies: 4
Views: 138

Re: import panel data

salhes wrote:
another question please, why I cannot import my data as in my first post, what is wrong so it does not accept it as panel.


fix the observations for Bolivia in the Excel file
by startz
Tue Jan 01, 2019 8:11 am
Forum: Data Manipulation
Topic: import panel data
Replies: 4
Views: 138

Re: import panel data

Note that you have two observations for Bolivia for 1960 and 1961. Probably the source of the problem.
by startz
Tue Jan 01, 2019 8:10 am
Forum: Data Manipulation
Topic: import panel data
Replies: 4
Views: 138

Re: import panel data - bug bug bug

It's worse than you think:

You can drag and drop the Excel file onto EViews. It will come in fine as an unstructured file. You can then use Proc/Structure to make a panel dated by series.

Unfortunately, this generates "Internal Error 500 500."
by startz
Sat Dec 15, 2018 12:35 pm
Forum: Estimation
Topic: F-test / Wald Test discrepancy
Replies: 4
Views: 195

Re: F-test / Wald Test discrepancy

I think your R^2 formula is wrong. Maybe the k-1 should be the number of restrictions, r
by startz
Thu Dec 13, 2018 6:39 am
Forum: Estimation
Topic: F-test / Wald Test discrepancy
Replies: 4
Views: 195

Re: F-test / Wald Test discrepancy

You'd have to post both sets of information for anyone to compare them.

As a wild guess, the sample changed when you did it manually.
by startz
Mon Dec 10, 2018 10:35 am
Forum: Bug Reports
Topic: Window refresh bug
Replies: 0
Views: 103

Window refresh bug

Here's a screenshot showing a refresh bug. Look at the row title on the cross-tab. What happened is that I dragged another window across this one and then dragged it off. Yes, I've done it twice with the same result.
Capture.PNG
Capture.PNG (27.82 KiB) Viewed 92 times
by startz
Thu Dec 06, 2018 7:46 am
Forum: Programming
Topic: Tables that contain series that sometimes aren't populated
Replies: 8
Views: 235

Re: Tables that contain series that sometimes aren't populated

One option--might or might not meet your needs--is to create the series and set them equal to NA. Then your code can populate them with real values or just leave them alone.
by startz
Wed Dec 05, 2018 7:33 pm
Forum: Estimation
Topic: "Log of non-positive number"
Replies: 16
Views: 13365

Re: "Log of non-positive number"

Logs can be negative, but the logs of a non-positive number don't exist.
by startz
Tue Dec 04, 2018 7:55 pm
Forum: Data Manipulation
Topic: Sample Survey workfile
Replies: 1
Views: 120

Re: Sample Survey workfile

What have you done so far?
by startz
Sun Dec 02, 2018 7:07 pm
Forum: Suggestions and Requests
Topic: @inlist delimitor
Replies: 0
Views: 205

@inlist delimitor

Maybe a version of @inlist which uses a comma for a delimitor? Not only would this be useful for people who don't RTFM and see you're supposed to use a space for a delimitor, it would also help for alpha series like "New York"
by startz
Fri Nov 30, 2018 2:59 pm
Forum: Data Manipulation
Topic: Shrinking workfile
Replies: 1
Views: 145

Shrinking workfile

This may be the dumbest question I've asked here in a while...

How do I save an EViews workfile while dropping observations not in the smpl? I have a 500MB workfile I need to shrink.
by startz
Thu Nov 22, 2018 10:02 am
Forum: Programming
Topic: Speeding up simulations
Replies: 5
Views: 1743

Re: Speeding up simulations

Be sure that under general options/advanced you have Compile to machine code on and multiprocessor use set to auto. But you probably do because these are the defaults. More RAM probably doesn't matter so long as you have enough to contain everything in memory. Unless you are doing a problem with an ...

Go to advanced search