 Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

# st: Re: st: Re: st: Matrices utilities, Basic calculations. When to start using Mata over Stata?‏

 From Tirthankar Chakravarty To statalist@hsphsun2.harvard.edu Subject st: Re: st: Re: st: Matrices utilities, Basic calculations. When to start using Mata over Stata?‏ Date Fri, 23 Apr 2010 20:01:42 +0530

```There is an error in my first batch of code. It should read:
******************************************
clear*
clear mata
forv i=1/20 {
matrix A`i' = matuniform(25,36)
matrix B`i' = matuniform(25,36)
matrix C`i' = J(25, 1, 0)
}

forv i=1/20 {
// need Ben Jann's -moremata- package
mata: mCtemp = mm_cond(rowsum(st_matrix("A`i'")):>
rowsum(st_matrix("B`i'")), rowsum(st_matrix("A`i'")),
rowsum(st_matrix("B`i'")))
mata: st_replacematrix("C`i'", mCtemp)
}
******************************************

T

2010/4/23 Austin Nichols <austinnichols@gmail.com>:
> You can loop in Stata and use Mata, or you can do rowsums by
> postmultiplying by a column vector of ones.
> e.g.
>
> forv n=1/10 {
>  mata: st_matrix("a`n'",runiform(25,36))
>  mata: st_matrix("b`n'",runiform(25,36))
>  mat c`n'=J(25,1,.)
> }
> mat li a10
> forv n=1/10 {
>  mata: a=st_matrix("a`n'")
>  mata: b=st_matrix("b`n'")
>  mata: ra=a*J(cols(a),1,1)
>  mata: rb=b*J(cols(a),1,1)
>  mata: c=ra:*(ra:>rb)
>  mata: st_matrix("c`n'",c)
> }
> forv n=1/10 {
>  mata: ra=rowsum(st_matrix("a`n'"))
>  mata: rb=rowsum(st_matrix("b`n'"))
>  mata: c=ra:*(ra:>rb)
>  mata: st_matrix("d`n'",c)
> }
> mat li c10
> mat li d10
>
>
>> Hi all,
>>
>> I have a total of 20 matrices, A1-A10 and B1-B10 of size
>> 25x36
>>
>> I am trying to do the sum of each row of the matrix A and
>> compare them with the sum of each row of the matrix B. Then I retain the
>>  highest sum and put it in a new matrix, say C.
>>
>> The "rowsum" utility in Mata is appealing, but I couldn't figure out how to create
>> the loops ( no "foreach" command?), generate new matrixes names
>> automatically (for the matrices C) and access my 20 matrices within the
>> loops in Mata.
>>
>> Here is what I would do if I could use "rowsum"
>> in Stata:
>>
>>
>> foreach n in 1 2 3 4 5 6 7 8 9 10 {;
>>
>>
>>  matrix c'n´= J(rowsof(A'n´),1,0)     *I create my matrix c (vector of
>> 25 rows) containing only 0 values ;
>>
>>     foreach i in 1-25 {;
>>         * where i correspond to the number of rows;
>>
>>
>>
>>  R_sum_A'n´=rowsum(A'n´) ;
>>         R_sum_B'n´=rowsum(B'n´) ;
>>
>>
>>
>>       *I compare the two sums and if the rowsum of A is bigger than B I
>> put it in the new matric C, if not I leave it equals to 0 ;
>>
>>
>>      if el(R_sum_A'n´,'i´,1)> el(R_sum_B'n´,'i´,1)
>>
>>
>>      matrix C'n´['i´,1] =el(R_sum_A'n´,'i´,1) ;
>>
>>
>>
>>
>>  };
>>  };
>>
>>
>> But I cannot use "rowsum" into stata . So what would you suggest ?
>>
>> Thanks a lot
>>
>> Raphael
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
>

--
To every ω-consistent recursive class κ of formulae there correspond
recursive class signs r, such that neither v Gen r nor Neg(v Gen r)
belongs to Flg(κ) (where v is the free variable of r).

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