Conformability Error Stata Matrix
Contents |
st: Conformability error in creating matrix while cross-tabulating data Date Thu, 5 Jul 2007 11:05:14 -0400 Dear Statalist, I am trying to get the frequency of certain variables and put it in a matrix (see stata conformability error r(503) code below). I need to cross tabulate each variable q271-q277 with each
Conformability Error Stata Factor
variable q256-q260. I then need to add the frequencies of q256-q258 as one group and q259-q260 as the other stata conformability error 503 group. Finally, I want to create a 7 x 2 matrix of the results (q271-q277 will be the 7 rows; and q256-q258 & q259-q260 will be the 2 columns). I ran
Conformability Error Mata
the following code, but am getting "conformability error". I think the problem is that I want a 7 x 2 matrix, but I'm using i = 271/277 rather than i=1/7. If this is the case, can you suggest an alternative? P.S. - Right now, I have variables q1-q7. I tried to drop them and rename q271-q277 as q1-q7 (in case this was stata display matrix the issue), without much luck. Thanks, Hugh matrix results = J(7,2,.); forval i = 271/277 {; count if q`i'~=. & q256~=.; local a = r(N); count if q`i'~=. & q257~=.; local a = `a' + r(N); count if q`i'~=. & q258~=.; matrix results[`i',1] = `a' + r(N); count if q`i'~=. & q259~=.; local b = r(N); count if q`i'~=. & q260~=.; matrix results[`i',2] = `b' + r(N); }; matrix colnames results = NP FP; matrix rownames results = q271 q272 q273 q274 q275 q276 q277; matrix li results, noheader; * * For searches and help try: * http://www.stata.com/support/faqs/res/findit.html * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/ Follow-Ups: st: RE: Conformability error in creating matrix while cross-tabulating data From: "Nick Cox"
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] Re: st: mata conformability error, but matrices are conformable From Misha Spisok
Matrix Stata
conformable Date Sat, 8 May 2010 15:09:14 -0700 Thank you, Professor Buis! This works
Stata Mata
(of course), but I have a further question about forcing a `noconstant' in the regression (I think I'll need e(b) to exclude a constant if my sketch continues as imagined). I have tried a number of things, but I cannot get it to work the way I'd like it to work. For example (same code with one change at regression http://www.stata.com/statalist/archive/2007-07/msg00165.html -poisson-, though I have the same problem with -regress-), program calcp, rclass version 10.1 syntax varlist [if] [in] marksample touse gettoken lhs rhs : varlist poisson `lhs' `rhs' if `touse', noconstant /* CHANGED: to try to force it to omit a constant */ mata: makexb("`rhs'", "`touse'") end mata: void makexb(string scalar rhs, string scalar touse) { real matrix X real rowvector b real colvector Xb real scalar idx X = st_data(., tokens(rhs), touse) X = http://www.stata.com/statalist/archive/2010-05/msg00412.html X, J(rows(X), 1, 1) b = st_matrix("e(b)") Xb = X*b' idx = st_addvar("float", "Xb") st_store(., idx, touse, Xb) } end I also tried changing the -syntax- line by including a non-optional noconstant option: , noconstant. When I try the above code, I get no constant (as desired), but I get this error: *: 3200 conformability error makexb(): - function returned error
many ways to R, Matlab or GAUSS. It does have some unique and intriguing features however. One is that it is a compiled language rather than interpreted, which improves performance. It also has been parallelized in Stata/MP (available on all the SSCC Linux servers and Condor) which dramatically improves performance. https://www.ssc.wisc.edu/sscc/pubs/4-26.htm On the other hand Mata is fairly new and has not yet "caught on" at the SSCC, so we don't have any real-world comparisons to offer. Mata is not a replacement for Stata, nor is it intended to be a stand-alone statistical package. It is a tool which is best used as a supplement to Stata (or SAS or SPSS), for doing those things Stata does not do well on its own. In particular, Mata does not work in the context of a conformability error single data set, giving it additional flexibility. But you should not try to learn Mata unless you are already familiar with Stata or another statistical package. Mata is a relatively "low level" language. Much of your time in Stata (or SAS or SPSS) is spent using built-in programs, finding just the right combination of options to get Stata to do what you want. In Mata you will take direct control, telling Mata what you want to do step-by-step. (The Mata optimizer, which we conformability error stata will discuss at length, is a notable exception.) That means doing simple things is usually more cumbersome in Mata than in Stata, but Mata has fewer constraints. This article is primarily written for people who have significant experience using Stata, SAS or SPSS syntax, but no other programming languages. Thus there will be a lot of emphasis on learning how to do useful things by manipulating matrices, and many of the examples are designed to give experience doing so as well as illustrating a particular concept. Matlab and GAUSS veterans may find they can skim these sections, focusing on what is new to them. C programmers will find that Mata imitates C whenever it can, so they can probably skim the sections on standard programming constructs like loops. But no matter what your background, you will learn far more if you read this article at the computer, with Stata running, and actually type in the examples. Mata runs within Stata, so in order to use Mata you'll need to know how to run a Stata program, called a do file. If you've never used Stata, please read the section on do files in An Introduction to Stata. Interactive Stata (i.e. start it up and type in commands) is a great way to learn and that's how you'll do the examples in this article. But for real work you'll want to write everything in do files. There are several example files associated with this article. There are
be down. Please try the request again. Your cache administrator is webmaster. Generated Wed, 05 Oct 2016 05:21:29 GMT by s_hv987 (squid/3.5.20)