Bookmark and Share

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


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

Re: st: Is there a better way to transpose a dataset and automatically label variables based on a string variable?


From   Nick Cox <[email protected]>
To   "[email protected]" <[email protected]>
Subject   Re: st: Is there a better way to transpose a dataset and automatically label variables based on a string variable?
Date   Fri, 21 Jun 2013 15:08:17 +0100

I am the author of the -sxpose- (SSC) referred to; it is as stated
designed to transpose a string variable dataset. What you did comes
with a warning, so I readily agree that it is not a good solution.

With your dataset, I can save the string values, -drop- that variable,
-xpose- and then copy back.

. l
     +---------------------------------------------------+
     |        varlabels   category   diff_c~y   catego~3 |
     |---------------------------------------------------|
  1. |       Some label     .79221     .67874     .70605 |
  2. |    Another label     .95949     .75774    .031833 |
  3. | Additional label     .65574     .74313     .27692 |
  4. |     Fourth label    .035712     .39223    .046171 |
  5. |      Fifth label     .84913     .65548    .097132 |
     |---------------------------------------------------|
  6. |      Final label     .93399     .17119     .82346 |
     +---------------------------------------------------+
. forval i = 1/6 {
  2. local label`i' = varlabels[`i']
  3. }
. drop varlabels
. xpose, clear
. l
     +--------------------------------------------------------+
     |     v1        v2       v3        v4        v5       v6 |
     |--------------------------------------------------------|
  1. | .79221    .95949   .65574   .035712    .84913   .93399 |
  2. | .67874    .75774   .74313    .39223    .65548   .17119 |
  3. | .70605   .031833   .27692   .046171   .097132   .82346 |
     +--------------------------------------------------------+
. forval i = 1/6 {
  2. label var v`i' "`label`i''"
  3. }

This could be programmed up for more general situations.
.
Nick
[email protected]


On 21 June 2013 14:40, John Bensin <[email protected]> wrote:
> It looks like my email client mangled my first post, and I think the
> formatting is relevant for the data sections, so I'll try this again.
>
> I have several datasets that fit this general format (variable names
> are in the first row):
>
>
> varlabels             category     diff_category     category3
> "Some label"        0.79221      0.67874         0.70605
> "Another label"     0.95949      0.75774         0.031833
> "Additional label"  0.65574      0.74313         0.27692
> "Fourth label"      0.035712     0.39223         0.046171
> "Fifth label"       0.84913      0.65548         0.097132
> "Final label"       0.93399      0.17119         0.82346
>
>
> My goal is to use Stata 12 to get them into this format:
>
> _var1      _var2      _var3      _var4      _var5      _var6        cat
> 0.792210   0.959490   0.655740   0.035712   0.849130   0.933990
> "category"
> 0.678740   0.757740   0.743130   0.392230   0.655480   0.171190
> "diff category"
> 0.706050   0.031833   0.276920   0.046171   0.097132   0.823460
> "category3"
>
> where _var1 is labelled "Some label", _var2 is labelled "Another
> label", etc. The datasets are different and numerous enough that
> labelling the variables individually isn't possible. The names of the
> category variables aren't predictable either.
>
> My current solution is to improvise with Mata, xpose, and destring/tostring
>
>
> clear
> input str50 varlabels category diff_category category3
> "Some label"        -18.2059     0.67874     105869.3
> "Another label"     0.95949      598.165     0.031833
> "Additional label"  0.65574      0.74313     0.27692
> "Fourth label"        194          0.39223     0.046171
> "Fifth label"       0.84913      394.069     -5968387
> "Final label"       0.93399      0.17119     0.82346
> end
>
> putmata varlabels, replace
>
> mata
>     varlabels = varlabels'
> end
>
> xpose, clear varname
> tostring v*, force usedisplayformat replace
> getmata (v*)=varlabels, update force
>
> foreach v of varlist v* {
>     local name = `v'[1]
>     lab var `v' "`name'"
> }
>
> drop if _varname == "varlabels"
> destring, replace
>
>
>
> Although this works, I question if there is a better way because I see
> a few problems with this:
>
> 1. The combination of tostring/destring may cause a loss of precision,
> according to Stata's warnings.
>
> 2. The --getmata-- call forces the vector of labels into the first
> observation, which seems clunky. Is it safe to always rely on this
> working or do I risk overwriting data?
>
> I also pursued a solution using the --sxpose-- command from the SSC:
>
>
> clear
> input str50 varlabels category diff_category category3
> "Some label"        0.79221      0.67874     0.70605
> "Another label"     0.95949      0.75774     0.031833
> "Additional label"  0.65574      0.74313     0.27692
> "Fourth label"        0.035712     0.39223     0.046171
> "Fifth label"       0.84913      0.65548     0.097132
> "Final label"       0.93399      0.17119     0.82346
> end
>
> sxpose, clear force
>
> foreach v of varlist _var* {
>     local name = `v'[1]
>     lab var `v' "`name'"
> }
>
>
> This costs me the names of the category variables, which I need to
> preserve, as well as some accuracy. It yields data that look like
> this:
>
> _var1             _var2               _var3               _var4
>        _var5            _var6
> Some label        Another label      Additional label     Fourth label
>      Fifth label        Final label
> 0.792209983        0.959490001        0.655740023        0.035712
>      0.849129975        0.933990002
> 0.678740025        0.757740021        0.743130028        0.392230004
>      0.655480027        0.171189994
> 0.706049979        0.031833001        0.276919991        0.046170998
>      0.097131997        0.823459983
>
>
> Any suggestions? I'm happy to continue with my Mata/xpose solution if
> it won't cost me data, but if there's a better solution I'll use it.
*
*   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index