Page 2 of 3

EqTabs (eqsumtab.prg edit)

Posted: Sat Aug 06, 2011 3:20 am
by patrix
I edited the eqsumtab.prg file, in order to display the significance asterixes as follows:
* for 10%
** for 5%
*** for 1%
Until now, the significance was shown as * for 5% and ** for 1%.
If it's not a problem I could attach it to a post on this forum.

Re: EqTabs (Equation summaries)

Posted: Sat Aug 06, 2011 8:53 am
by EViews Gareth
Sure, others might appreciate you posting it.

Re: EqTabs (Equation summaries)

Posted: Sat Aug 06, 2011 11:37 pm
by patrix
Here is the file. I think the changes are ok, but you may check.

Re: EqTabs (Equation summaries)

Posted: Sun Sep 23, 2012 7:45 pm
by kat15
Hi there,

I estimated about 60 GARCH equations, and using the EqSum add-in to summarise all the coefficient reuslts in one place ... However, I am not getting any F-statistics (says "N/A" for each equation) ... any idea how to fix this?

Also, I was wondering whether it would be possible to add a Wald test into the summary output?? Can I add this into the program code at all? Would be very helpful, as I need to do a coefficient restriction test for each equation, and I really don't want to do it manually about 300 times!

Thank you! :)

Re: EqTabs (Equation summaries)

Posted: Sun Sep 23, 2012 10:25 pm
by EViews Gareth
EViews doesn't calculate the F statistic for ARCH equations.

If you want to add the Wald statistics, you'll have to program it yourself, unfortunately.

Re: EqTabs (Equation summaries)

Posted: Fri Apr 05, 2013 12:39 am
by eviewsuser77

I have created a workfile which contains a set of variables and a set of equations that have been saved for a set of 41 companies. My aim is to be able to transfer the outputs of the equations (several specifications for each company, >100 here even accounting for the nonvalid ones and further additions possible) not manually one by one by copying and pasting in Excel, but to kind of create one large table with all the estimated equations inside (I need coefficients, t-stats and P-values at least) or several such tables that would together cover all the equations and could be transferred to Excel.

After a thorough search for some helpful tool, I found the EqTabs add-in, I read everything I could find about it but so far I haven't managed to successfully use it. A description of my effort to use it, after downloading it and opening with EViews 6, together with sreenshots, is attached to the email as a MS Word file. I would appreciate your help in either enabling me to use this tool to achieve my aim, or any other method that would help achieve my aim. Doing the operations described above manually is not an option for me since the number of equations is expected to increase and some more automatic way would be the only thing that could save me:-). I can send the workfile as well if that's helpful

Thank you very much in advance!

Re: EqTabs (Equation summaries)

Posted: Fri Apr 05, 2013 8:20 am
by EViews Gareth
All EViews add-ins require EViews 7.1 or above, and cannot be used with the Student Version.

Re: EqTabs (Equation summaries)

Posted: Mon Apr 15, 2013 7:17 am
by Novi
Sorry 4 posting with delay about this but why does the add-in not include all the independent variables in the equations but just the first one? is it only happening for me? I have re-installed the add-in just today but I don't seem to get any improvement on this :(


Re: EqTabs (Equation summaries)

Posted: Mon Apr 15, 2013 8:12 am
by EViews Gareth
Could you explain in more detail? I don't understand what you're saying.

Re: EqTabs (Equation summaries)

Posted: Mon Apr 15, 2013 8:16 am
by Novi
Sure I can, sorry!

I want to create a stacked table summarizing the results of 5 equations, each of which has 3 explanatory variables...nevertheless by using the add-in, the stacked table I get reports the coefficient for the first explanatory variable only, neglecting the remaning two.

Is it clearer now? it only seems to be a bug in the add-in or something but maybe it is not so difficult to fix...

Re: EqTabs (Equation summaries)

Posted: Mon Apr 15, 2013 8:21 am
by EViews Gareth
Must be something specific to your workfile/equations. I just ran it, and it listed all regressors.

Re: EqTabs (Equation summaries)

Posted: Sun Apr 21, 2013 1:33 pm
by dagfinnrime

Thanks. This is a very handy add-in.

I have one comment and one requests:
1) use @rbar2 instead of @r2. (of course, users can easily fix this)
2) Add a "transposed" version. Columns are variables, obs and R2; rows are equations (with std.error or t-stats below)


Re: EqTabs (Equation summaries)

Posted: Wed May 08, 2013 7:36 pm
by JimForest
I have attached a file that contains a modified version of the eqsumtab.prg add-in program.

The modified program produces Adjusted R-Squared, instead of R-Squared.

The only modifications made are to the label in the GUI, which now reads 'Adj. R-Squared', and to the called “@-function” is now @rbar2 instead of @r2.

This is just a personal preference of mine for output tables.

Re: EqTabs (Equation summaries)

Posted: Thu Apr 10, 2014 1:01 pm
by dreavis
Can I add add forecast and residual series to the Equation Summary program show it will show in the equation columns, below the results and R-square? So that a single column contains an Equation Result, then the Forecast Series, and then the Residuals Series below. These are forecasts and residual series were created from the equation using .forecast and .makeresids, so they are kind of like children of the equation. I thought perhaps I could use an associated data member, but I do not see that as an option in the Series Data Members, and I don't see an Equation Data Member, but I could be missing it. I am attaching a (really rough) mock-up of what I have in mind in case it will help paint a picture of what I want to do. Any advice would be appreciated!

I have only worked with EViews for about 6 months but I love it.

I made two other changes to the Eq Summ program - I added the p-values to the coefficients (listed under the t-stat), added a 10-percent significance level (I did notice somebody else did that a couple of years ago :wink: ), and I added the DW stat below the R-square.

Re: EqTabs (Equation summaries)

Posted: Thu Apr 10, 2014 1:18 pm
by EViews Gareth
You'd have to program it yourself. The reason that forecast and residuals aren't datamembers is that they require extra calculations, and options. They are not stored with the equation by default, but are recalculated every time you do .forecast or makeresid.

What you'll have to do is add a line that forecasts from each equation (eq.forecast yf), then freeze the sheet view of the resultant forecast series into a table, then extract that table into the summary table. The only tricky bit will be getting the sizing/placement of the table correct. Finicky.