RE: st: One varlist into a matrix with multiple columns

[Joe Canner <jcanner1@jhmi.edu>]

But, Scott's suggestion using -reshape- could be followed by the following Mata command (or similar):

a=st_data(.,"x1-x10")

which I think would achieve the desired result and be a little easier to understand.  Of course, -reshape- may not be the best option if the matrix is sparse or the indices irregular.

Joe Canner

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Sergiy Radyakin
Sent: Wednesday, August 14, 2013 5:01 PM
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: One varlist into a matrix with multiple columns

No he wants a matrix:

clear
set obs 10
gen i = _n
expand 10
bys i: gen j = _n
gen x = rnormal()

*** this is only one possible way of determining the matrix size **** sum i local rows=r(max) sum j local cols=r(max) matrix M=J(`rows',`cols',.)
*********************************************************************

forval z=1/`=_N' {
  local tr=i[`z']
  local tc=j[`z']
  matrix M[`tr',`tc']=x[`z']
}

matrix list M


On Wed, Aug 14, 2013 at 4:48 PM, Scott Merryman <scott.merryman@gmail.com> wrote:
> Do you mean something like:
>
> clear
> set obs 10
> gen i = _n
> expand 10
> bys i: gen j = _n
> gen x = rnormal()
> reshape wide x , i(i) j(j)
> l
>
> Scott
>
>
> On Wed, Aug 14, 2013 at 3:15 PM, Samuel Bailey <smbailey1@uchicago.edu> wrote:
>> Hello,
>>
>> I have a dataset with three columns: the first two are the i and j 
>> coordinates of a potential matrix, and the third is the data that 
>> would go into the (i,j)th spot. Is there a way (in Stata or Mata) to 
>> create a matrix directly from these three variables? I've tried using
>> -mkmat- to create a column vector which gets sent to Mata, where it's 
>> looped through into a new matrix. However, as a column vector the 
>> data exceeds the maximum matrix size in my version of Stata (though 
>> as a square matrix it would not).
>>
>> Sincerely,
>> Sam Bailey
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/