Looping using list of var with similar prefix

[Alexandre Kaisin]

Hi everyone!

I'm getting stuck with something probably silly.

I have a number of variables that share a common prefix which I would like to regress individually using reghdfe.
What I have come up with is not working. I'm getting an 'unexpected end of file'.

foreach var of spei_* {
    quietly reghdfe ln_agprod var, absorb(country_id continent_id#year) vce(cluster country_id)
    estimates store est_`var'
}

My variables are of the form spei_n20 to spei_n0 and spei_1 to spei_20.

Many thanks for your help!

[Carlo Lazzaro]

Alexandre:
provided that it's not a trivial typo, the first thing that I would amend in the second line of your code is:

`var'

instead of:

var

The following toy-example may help:

set obs 100
number of observations (_N) was 0, now 100

. g y=runiform()*1000

. g Carlo_A=runiform()*100

. g Carlo_B=runiform()*10

. g Carlo_C=runiform()*20

. foreach var of varlist Carlo_* {
  2. regress y `var'
  3. estimates store `var'
  4.  }

      Source |       SS           df       MS      Number of obs   =       100
-------------+----------------------------------   F(1, 98)        =      0.11
       Model |  10846.5994         1  10846.5994   Prob > F        =    0.7377
    Residual |  9422817.35        98  96151.1974   R-squared       =    0.0011
-------------+----------------------------------   Adj R-squared   =   -0.0090
       Total |  9433663.95        99  95289.5348   Root MSE        =    310.08

------------------------------------------------------------------------------
           y |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
     Carlo_A |   .3687117   1.097785     0.34   0.738    -1.809808    2.547231
       _cons |   478.5269   64.02046     7.47   0.000     351.4804    605.5735
------------------------------------------------------------------------------

      Source |       SS           df       MS      Number of obs   =       100
-------------+----------------------------------   F(1, 98)        =      0.01
       Model |  664.331479         1  664.331479   Prob > F        =    0.9340
    Residual |  9432999.62        98  96255.0981   R-squared       =    0.0001
-------------+----------------------------------   Adj R-squared   =   -0.0101
       Total |  9433663.95        99  95289.5348   Root MSE        =    310.25

------------------------------------------------------------------------------
           y |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
     Carlo_B |   .8930472   10.74964     0.08   0.934    -20.43926    22.22536
       _cons |   492.4435   66.59521     7.39   0.000     360.2874    624.5995
------------------------------------------------------------------------------

      Source |       SS           df       MS      Number of obs   =       100
-------------+----------------------------------   F(1, 98)        =      2.26
       Model |  212690.526         1  212690.526   Prob > F        =    0.1359
    Residual |  9220973.42        98  94091.5655   R-squared       =    0.0225
-------------+----------------------------------   Adj R-squared   =    0.0126
       Total |  9433663.95        99  95289.5348   Root MSE        =    306.74

------------------------------------------------------------------------------
           y |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
     Carlo_C |  -8.099759   5.387328    -1.50   0.136    -18.79074    2.591217
       _cons |   577.8113    61.6906     9.37   0.000     455.3883    700.2343
------------------------------------------------------------------------------

[Alexandre Kaisin]

thanks you so much.
I was too quick with posting a question. I figured it out.
Is there a way to delete the post?

[Carlo Lazzaro]

Alexandre:
you can edit your post within 1 hour from posting it.
But there's nothing sinister in leaving in a post about which you have a second thought, as your original query may be useful to other listers.
That said, if you have other questions, feel free to post them under a meaningful title. Thanks.

[Nick Cox]

All this is explained in the FAQ all posters are asked to read before posting. See https://www.statalist.org/forums/help#closure

You cannot delete a post.

Naturally, there are circumstances in which the OP regrets a post. And usually that arises from modesty and/or embarrassment. We all ask silly questions sometimes.

But the bigger deal is that on Statalist an OP has no rights here over postings, which belong to the forum -- certainly no rights over the posts of anyone who replies. More positively, the ideal is that other people may find your post interesting or helpful. If one person can ask a question, another person may have the same question.

In one place I contribute to, the OP can delete -- well, hide -- an entire thread, including lengthy replies intended to help. In practice that means, most often, students with assignments covering their tracks -- sometimes even to stop help being visible to their classmates. Long discussion possible, but that's some way from the ethos we want here.