Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: RE: Save -table- command results


From   Nick Cox <njcoxstata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: RE: Save -table- command results
Date   Tue, 13 Mar 2012 01:06:12 +0000

I'll finish what I started. Weights now supported.

N.B. This is just a demonstration for Pablo of the principle that you
can write your own program. It stops here unless anyone else wants the
code for their own.

Examples:

 sysuse auto
(1978 Automobile Data)

. mysums turn trunk mpg, by(rep78)

. matrix li mysums

mysums[5,3]
    turn  trunk    mpg
1     82     17     42
2    347    117    153
3   1232    458    583
4    693    243    390
5    392    126    301

. mysums turn trunk mpg, by(rep78) matname(different)

. mat li different

different[5,3]
    turn  trunk    mpg
1     82     17     42
2    347    117    153
3   1232    458    583
4    693    243    390
5    392    126    301

. mysums turn trunk mpg if foreign, by(rep78) matname(different)

. mat li different

different[3,3]
    turn  trunk    mpg
3    106     37     70
4    315     93    224
5    320    107    237


program mysums
       version 8.2
       syntax varlist(numeric) [if] [in] [aweight fweight iweight],
by(varname) [matname(name)]

       quietly {
               marksample touse
               markout `touse' `by', strok
               count if `touse'
               if r(N) == 0 error 2000

               local J : word count `varlist'
               tempvar group
               egen `group' = group(`by') if `touse'
               su `group', meanonly
               local I = r(max)

               if "`matname'" == "" local matname "mysums"
               matrix `matname' = J(`I', `J', .)
               tokenize `varlist'

               forval i = 1/`I' {
                       forval j = 1/`J' {
                               su ``j'' if `group' == `i' [`weight'
`exp'], meanonly
                               matrix `matname'[`i', `j'] = r(sum)
                       }
               }

               levelsof `by' if `touse'
               capture matrix rownames `matname' = `r(levels)'
               matrix colnames `matname' = `varlist'
       }
end

On Mon, Mar 12, 2012 at 3:53 PM, Nick Cox <n.j.cox@durham.ac.uk> wrote:
> At this distance -statsmat- doesn't seem a very interesting command to change. Kit may be more enthusiastic.
>
> It's perhaps more instructive to show how to write your own program. I didn't add in weights, but that's not difficult, and I have a meeting somewhere else in a few minutes' time.
>
> program mysums
>        version 8.2
>        syntax varlist(numeric) [if] [in] , by(varname) [matname(name)]
>
>        quietly {
>                marksample touse
>                markout `touse' `by', strok
>                count if `touse'
>                if r(N) == 0 error 2000
>
>                local J : word count `varlist'
>                tempvar group
>                egen `group' = group(`by') if `touse'
>                su `group', meanonly
>                local I = r(max)
>
>                if "`matname'" == "" local matname "mysums"
>                matrix `matname' = J(`I', `J', 1)
>                tokenize `varlist'
>
>                forval i = 1/`I' {
>                        forval j = 1/`J' {
>                                su ``j'' if `group' == `i', meanonly
>                                matrix `matname'[`i', `j'] = r(sum)
>                        }
>                }
>
>                levelsof `by' if `touse'
>                capture matrix rownames `matname' = `r(levels)'
>                matrix colnames `matname' = `varlist'
>        }
> end
>
> Nick
> n.j.cox@durham.ac.uk
>
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Pablo Bonilla
> Sent: 12 March 2012 15:33
> To: Stata List
> Subject: Re: st: RE: Save -table- command results
>
> Dear Nick,
>
> You'r right, I am sorry for not explain the provenance os -statsmat-.
>
> Thank you for the suggestion, but still it is faster and more efficient to use the -statsmat- command  because the preserver and the collapse commands are very slow in huge data bases like the one I am using.
>
> That would be great if the -statsmat- command can present results with the by() option for several variables.
>
> Best,
>
> Pablo.
> Sent via BlackBerry from T-Mobile
>
> -----Original Message-----
> From: Nick Cox <n.j.cox@durham.ac.uk>
> Sender: owner-statalist@hsphsun2.harvard.edu
> Date: Mon, 12 Mar 2012 14:58:02
> To: 'statalist@hsphsun2.harvard.edu'<statalist@hsphsun2.harvard.edu>
> Reply-To: statalist@hsphsun2.harvard.eduSubject: st: RE: Save -table- command results
>
> -statsmat- is from SSC (Christopher F. Baum and friend). Please remember to explain the provenance of user-written commands.
>
> You don't have to use -statsmat- at all to get a matrix. There are many ways of doing it.
>
> Here is one:
>
> preserve
> collapse (sum) var1 var2 var3 var4 var5 [w=factor], by(decil)
> mkmat var*, matrix(mysums) row(decil)
> restore
>
> Nick
> n.j.cox@durham.ac.uk
>
> Pablo Bonilla
>
> I would like to know if there is a way to save the results from the
> -table- command in matrices. I am using the -statsmat- command but It
> is very inefficient if I need to use it with the by() option.
>
> For instance, this is what I want to save:
> table decil [w = factor], c(sum var1  sum var2 sum var3  sum var4 sum
> var5)  f(%19.2f) row
>
> But I have to do it in this way in order to get a matrix:
> cap mat drop mat1
> foreach var in var1 var2 var3 var4 var5 {
>        statsmat `var' [w = factor], by(decile) stat(sum) format(%16.0f)
> matrix(`income')
>        mat mat1 = nullmat(INCOME),`income'
> }
>

*
*   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/


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