There are two interpretations to that question. The first is "what is the mathematical way of defining the troughs", the second is "How do I do that in EViews".
I can't figure out which one you're asking.
Search found 13318 matches
- Fri Nov 05, 2010 11:38 am
- Forum: Data Manipulation
- Topic: Recession analysis
- Replies: 4
- Views: 4156
- Fri Nov 05, 2010 8:48 am
- Forum: Programming
- Topic: Finding common last observation in workfile
- Replies: 6
- Views: 5988
Re: Finding common last observation in workfile
Ah, yes, I misunderstood whether you wanted the last date for which any series has a value or the last date for which all series have a value.
- Fri Nov 05, 2010 7:17 am
- Forum: Programming
- Topic: Finding common last observation in workfile
- Replies: 6
- Views: 5988
Re: Finding common last observation in workfile
I'm posting from my phone, so can't test this, but something along the lines of:
Code: Select all
smpl if @robs(g)>0
series t = @trend+1
string date = @otod(@max(t))
- Thu Nov 04, 2010 6:26 pm
- Forum: Programming
- Topic: Log of non positive number
- Replies: 1
- Views: 2705
Re: Log of non positive number
That's a crazy looking loop you have there, it seems you're running two loops, one inside the other, using !i for both. However I don't think that is causing the error. The most likely cause is that one of the ARCH estimations is going wrong. Nobody is going to be able to help much more than that wi...
- Thu Nov 04, 2010 11:51 am
- Forum: General Information and Tips and Tricks
- Topic: Database, fetch or copy commands
- Replies: 2
- Views: 4601
Re: Database, fetch or copy commands
I think you'll have to go through a workfile. Copy from the database to a workfile of the desired frequency, then copy back to a database.
- Thu Nov 04, 2010 10:56 am
- Forum: Add-in Support
- Topic: VARForecast (VAR forecasting)
- Replies: 44
- Views: 69309
Re: VARForecast (VAR forecasting)
Hadn't even occurred to me what would happen to the display names - I'll make it something better. Any suggestions on what it should be?
- Wed Nov 03, 2010 11:04 am
- Forum: Installation and Registration
- Topic: Update bug
- Replies: 2
- Views: 6093
Re: Update bug
New one should be up.
- Wed Nov 03, 2010 10:31 am
- Forum: Installation and Registration
- Topic: Update bug
- Replies: 2
- Views: 6093
Re: Update bug
Our mistake. We'll put up a new one soon.
- Wed Nov 03, 2010 10:17 am
- Forum: Programming
- Topic: Monte Carlo simulation
- Replies: 1
- Views: 5142
Re: Monte Carlo simulation
A very rough, quick version: !t = 100 'workfile size !k = 3 'number of variables !l = 2 'number of lags in VAR !its = 100 'number of repititions. 'make workfile create u !t 'declare VAR object to be used for estimation VAR theVAR 'make string containing names of variables (populated in the loop belo...
- Tue Nov 02, 2010 8:45 pm
- Forum: Add-in Support
- Topic: RecShade (add US recession shading)
- Replies: 27
- Views: 75470
Re: RecShade (add US recession shading)
If you open the Add-in program, you'll notice there's a ton of lines that say
Change those all to be
to change the colour to green, etc...
Code: Select all
g1.draw(shade, bottom)
Change those all to be
Code: Select all
g1.draw(shade, bottom, color(green))
to change the colour to green, etc...
- Tue Nov 02, 2010 8:03 am
- Forum: Programming
- Topic: a simulation on AIC model selection
- Replies: 1
- Views: 2599
Re: a simulation on AIC model selection
I just ran your code, and although it often chose 1 lag, it often didn't (i.e. it chose other numbers). I can't see anything really wrong with your code, to be honest.
- Tue Nov 02, 2010 7:56 am
- Forum: Estimation
- Topic: principal component factor analysis
- Replies: 13
- Views: 16777
Re: principal component factor analysis
You can use the PCOMP group view/proc. It is found in the Object Reference in the Group section.
- Tue Nov 02, 2010 7:53 am
- Forum: Data Manipulation
- Topic: How to put a correlogram in word?
- Replies: 2
- Views: 6244
Re: How to put a correlogram in word?
The only real way is to make a screenshot (press print-screen, then paste in Word)
- Mon Nov 01, 2010 1:52 pm
- Forum: Econometric Discussions
- Topic: How to interpret results of the Empirical Distribution Test?
- Replies: 4
- Views: 9685
Re: How to interpret results of the Empirical Distribution Test?
Without getting too technical, if you run your data against 100 different empirical distributions, at 95% significance, you'll come up with 5 that say they match, even though they don't. In general you can't run multiple econometric/statistical tests in a row without adjusting for the fact that you ...
- Mon Nov 01, 2010 1:50 pm
- Forum: Suggestions and Requests
- Topic: Custom Functions
- Replies: 1
- Views: 3397
Re: Custom Functions
Not for the forseeable future - it is really hard for us to implement. Although we'd love to do it, allocating the time when there are so many other cool features... Of course you can sort of use an Add-in that way. Write a program that does what you want, then give it a proc name, plus make it take...