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

Re: st: Re: basic matrix definition question

From   "Thomas Jacobs" <>
Subject   Re: st: Re: basic matrix definition question
Date   Tue, 16 Oct 2007 09:16:09 -0500

Dear Kit,

I found this out the hard way spending most of the day coding in Stata
to get a single observation working and then choking when I discovered
(OK, rediscovered) the 11,000 row limit in a Stata/SE matrix would
prevent me from getting a matrix loop working as my ultimate target
dataset exceeds 1 million rows.  I then re-coded the whole thing in
Mata in very little time.  I found the language much more intuitive
than stata programming (no gen vs. macro decisions or logic just set X
= whatever and move on).  As I said in another post, I will look
forward to trying the maximize as you suggested.  I have found the
lack of examples in the Mata documentation pretty frustrating as I try
to move my results back to Stata.  More on this in an upcoming post.
Thanks again.



On 10/16/07, Kit Baum <> wrote:
> Doing something like this in Stata's matrix language is IMHO an
> exercise in frustration (and punctuation). Per my recent posting on
> solving nonlinear equations, you would be much better off setting up
> the problem in Mata, making use of -optimize-.
> Kit Baum, Boston College Economics and DIW Berlin
> An Introduction to Modern Econometrics Using Stata:
> On Oct 16, 2007, at 2:33 AM, statalist-digest wrote:
> > Thanks very much.  I totally missed the mkmat command in the manual.
> > Although I probably made the example too simple, I am trying to write
> > a program to solve for two equations in two unknowns so I am building
> > a partial derivative matrix for inverting.  I calculate each of the 4
> > partials as variables and then combine them into a matrix.  I was
> > stuck on the create matrix step.  Using local macros worked as well
> > with a matrix command.  I assume that I would need to first create two
> > column vectors from 2 of the four variables and then combine these
> > with another mkmat command.
> *
> *   For searches and help try:
> *
> *
> *

Thomas Jacobs
*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index