[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Small Stata for teaching

From   Keith Dear <>
Subject   Re: st: Small Stata for teaching
Date   Wed, 15 Aug 2007 09:09:23 +1000

I too had a frustrating experience with Small Stata recently. A student could not -insheet- his dataset, and it was hard to see why not: no declared limit was being breached (34 variables; 42 rows; only 13kb) and my installation (Intercooled 9.2) accepted it without a murmur.

See -help limits- (it took us half a day to find this). There is an additional restriction in Small Stata: the maximum WIDTH of the data cannot exceed 200 bytes. His data included a str150 comment field: omit that and it worked fine (the width then was still 164: he had some other strings).

The error message explained "Stata can trade off width and length", but with this is evidently true only up to a point, the point for Small Stata being a width of just 200 bytes.

-h datatypes- gives the widths of the various types of variable. You can find the width of a dataset (once loaded!) using -describe- which returns r(width) and, I now see, r(widthmax) which I suppose for Small Stata would show 200 (for me, r(widthmax) is 24576 and -h limits- says 24564, CEFGW).


At 02:07 AM 15/08/2007, you wrote:

Dear Statalisters,

I am sure some of you have taught statistics/econometrics courses using Stata and recommended that students buy Small Stata.
The advertised limitations for Small Stata do not seem too restrictive: a maximum of 99 variables, 39 right-hand variables, 1000 observations and a matsize of 40.
The course I am teaching is on economic forecasting. I am using the book
"Elements of Forecasting" by F. Diebold and contrary to the examples on the book (which are on Eviews) I decided to use Stata. Unfortunately, I am finding out that the students can not replicate some of the examples on the book because of the restrictions on Small Stata. For example, they are unable to estimate an ARMAX with 12 seasonal variables using 400 observations. From the general description of Small Stata it never crossed my mind that they couldn't do it. This has been a frustrating experience as I am never sure what exactly are the limitations of Small Stata. I definitely would expect it to be able to run most "textbook examples" and I naively assumed that to be true.
For the next time that I teach the course I have to decide if I change the book or the software. I would like to know about your experiences
with Small Stata and if this also happens with other textbooks.

Paulo Guimaraes
* For searches and help try:

Dr Keith B.G. Dear
Senior Fellow in Biostatistics
National Centre for Epidemiology and Population Health
Australian National University
Canberra, ACT 0200, Australia
Tel: 02 612 54865, Fax: 02 612 50740
CRICOS provider #00120C

*   For searches and help try:

© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index