The RATS Software Forum

You've probably already figured this out, but based on your description of the file, it appears that you have 4 columns of information, but only 2 of those columns have series names. If you add "series" names for the first two columns, such as "name" and "year", the err...

Thanks to Todd for emailing the files. You can download them using the following links.

Data file:
www.estima.com/procs/clark_uhlig/data_q.xls

Program file:
www.estima.com/procs/clark_uhlig/UhligBVAR.prg

You can find the data files on Ruey Tsay's web site. The link in our "textbook examples" section should take you there.

Or order RATS 6.35, which will be shipping in a few days.

Todd:

If you like, I can store a copy of it on our website and post a link here. Just email me a copy as an attachment if you want to do that.

Thanks,
Tom Maycock
support@estima.com

This is an example which uses MVGARCHFORE in a non-trivial fashion. Note that because "h" is a SERIES[SYMMETRIC], you reference its elements as h(time)(i,j). This also demonstrates the univariate forecasting procedure, GARCHFORE . The example is from Ruey Tsay's Analysis of Financial Time ...

Hi Tom Many thanks for your answer. - "split graphs", I will send you an example today via email Thanks, as noted in my email: You can already do this to some extent using the "OVRANGE" option (in conjunction with OVERLAY and related options) on GRAPH. You specify a fraction ind...

Alex: Thanks for the suggestions. Some of these are already in the works. I did have a couple of questions. First, note that you can already choose the color used for lines/patterns from a set of predefined choices by using the "symbol choice" parameter on the supplementary card(s). Is thi...

Sorry, didn't see this post earlier. The PFORM instruction can do this easily. The easiest approach would be to change your company codes from letters to series, so RATS can process them as series. Then, just do a single PFORM command per series to reorder the data. For example: allocate 6 data(unit...

Kai Carstensen from the IFW in Kiel sent us the procedures and example program for his article "Stock Market Downswing and the Stability of European Monetary Union Money Demand", Journal of Business & Economic Statistics 2006 (October) 395-402. The main procedure implements Fully Modif...

I think this is pretty well covered in Section 14.3 of the User's Guide (Censored and Truncated Samples), and the LDV section of the Reference Manual.

Do you mean literally a variable of type: SERIES[SYMMETRIC] If so, the easiest way is probably to copy the data into a set of regular series and then use COPY. You could do it one at a time: set ss11 = ss(t)(1,1) set ss21 = ss(t)(2,1) etc. but an easier way is to create a [SYMMETRIC[SERIES]]. Assumi...

Gregory:

If you're still having problems with this, email a copy of the program and data file along with your RATS version number and serial number to support@estima.com and we'll take a look.

Tom Maycock

In the old days, one had to use specific commands for real computations (EVAL), integer computations (IEVAL), and matrix computations (MATRIX). Those were all superseded by the COMPUTE instruction, which I believe was introduced in Version 4. The old commands are still supported, but for any new cod...

Dear All, I am estimating a SURE model with inst option. My variable is in first difference. I need to get forecast for the variable in levels, but the model is estimated in first difference for unit root issues. Is there a way to convert the forecasted values (both in sample and out of sample) fro...

The problem is that you are doing integer division here: 2/LAGS Integer division returns an integer result, with any remainder truncated. So, if LAGS = 3, you are getting: 2/3 = 0.66667 which truncates down to zero. Assuming you want a fractional result, you need to make one or both terms real rathe...