Bug? Regress with absorb does not store r(table)

Mauricio Caceres

Join Date: Sep 2015

Posts: 135
#1

Bug? Regress with absorb does not store r(table)

21 Nov 2025, 03:09

In Stata 19.5, regress with absorb() does not store r(table). Instead, it lists r(datasignature).

Code:

sysuse auto, clear reg price mpg, absorb(rep78) return list

Workaround is to just replay the table, but the behavior looks like a bug? The table does get stored without absorb(), and areg does store r(table) as well.
Tags: None

1 like

Richard Williams

Join Date: Apr 2014
Posts: 5044

21 Nov 2025, 08:20

I have the same problem in 19.5. But in 18.5,

Code:

. version
version 18.5

. reg price mpg, absorb(rep78)

Linear regression, absorbing indicators         Number of obs     =         69
                                                F(1, 63)          =      20.72
                                                Prob > F          =     0.0000
                                                R-squared         =     0.2584
                                                Adj R-squared     =     0.1995
                                                Root MSE          =     2605.8

------------------------------------------------------------------------------
       price | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
         mpg |  -280.2615   61.57666    -4.55   0.000    -403.3126   -157.2103
       _cons |   12112.77   1347.968     8.99   0.000      9419.07    14806.47
------------------------------------------------------------------------------

. return list

matrices:
              r(table) :  9 x 2

. mat list r(table)

r(table)[9,2]
               mpg       _cons
     b  -280.26149    12112.77
    se   61.576661   1347.9683
     t  -4.5514239   8.9859455
pvalue    .0000249   6.848e-13
    ll  -403.31262   9419.0701
    ul  -157.21035    14806.47
    df          63          63
  crit   1.9983405   1.9983405
 eform           0           0

So, yes, it sounds like a bug to me.

-------------------------------------------
Richard Williams
Professor Emeritus of Sociology
University of Notre Dame
StataNow Version: 19.5 MP (2 processor)
EMAIL: [email protected]
WWW: https://academicweb.nd.edu/~rwilliam/

Comment

Andrew Musau

Join Date: Oct 2014

Posts: 10503
#3

21 Nov 2025, 14:27

Originally posted by Mauricio Caceres View Post

In Stata 19.5, regress with absorb() does not store r(table). Instead, it lists r(datasignature).

Code:

sysuse auto, clear reg price mpg, absorb(rep78) return list

Workaround is to just replay the table, but the behavior looks like a bug? The table does get stored without absorb(), and areg does store r(table) as well.

The -absorb()- option of regress is undocumented, so I doubt you will get official support for it. Among other issues, as this thread shows, it produces incorrect residuals after estimation. I would suggest using areg instead, as it is the official linear regression estimator that absorbs indicators.
2 likes
Comment

Carlo Lazzaro

Join Date: Apr 2014
Posts: 17859

22 Nov 2025, 01:32

Dear All,
I cannot replicate Mauricio's problem with Stata 19.0 upgraded to Stata 19.5:

Code:

. sysuse auto, clear
(1978 automobile data)

. 
. reg price mpg, absorb(rep78)

Linear regression, absorbing indicators         Number of obs     =         69
                                                F(1, 63)          =      20.72
                                                Prob > F          =     0.0000
                                                R-squared         =     0.2584
                                                Adj R-squared     =     0.1995
                                                Root MSE          =     2605.8

------------------------------------------------------------------------------
       price | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
         mpg |  -280.2615   61.57666    -4.55   0.000    -403.3126   -157.2103
       _cons |   12112.77   1347.968     8.99   0.000      9419.07    14806.47
------------------------------------------------------------------------------

. 
. return list

matrices:
              r(table) :  9 x 2

. mat list r(table)

r(table)[9,2]
               mpg       _cons
     b  -280.26149    12112.77
    se   61.576661   1347.9683
     t  -4.5514239   8.9859455
pvalue    .0000249   6.848e-13
    ll  -403.31262   9419.0701
    ul  -157.21035    14806.47
    df          63          63
  crit   1.9983405   1.9983405
 eform           0           0


. areg price mpg, absorb(rep78)

Linear regression, absorbing indicators          Number of obs     =        69
Absorbed variable: rep78                         No. of categories =         5
                                                 F(1, 63)          =     20.72
                                                 Prob > F          =    0.0000
                                                 R-squared         =    0.2584
                                                 Adj R-squared     =    0.1995
                                                 Root MSE          = 2605.7822

------------------------------------------------------------------------------
       price | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
         mpg |  -280.2615   61.57666    -4.55   0.000    -403.3126   -157.2103
       _cons |   12112.77   1347.968     8.99   0.000      9419.07    14806.47
------------------------------------------------------------------------------
F test of absorbed indicators: F(4, 63) = 1.072               Prob > F = 0.378

. return list

matrices:
              r(table) :  9 x 2

. mat list r(table)

r(table)[9,2]
               mpg       _cons
     b  -280.26149    12112.77
    se   61.576661   1347.9683
     t  -4.5514239   8.9859455
pvalue    .0000249   6.848e-13
    ll  -403.31262   9419.0701
    ul  -157.21035    14806.47
    df          63          63
  crit   1.9983405   1.9983405
 eform           0           0

.

Kind regards,
Carlo
(Stata 19.0)

Comment

Richard Williams

Join Date: Apr 2014

Posts: 5044
#5

22 Nov 2025, 06:31

Carlo Lazzaro , odd that I can replicate in Stata 19.5 but you can't. When did you last update Stata? I wonder if something got changed in a maintenance release? Also I am on a Win11 machine.

Code:

. update query (contacting https://www.stata.com) Update status Last check for updates: 22 Nov 2025 New update available: none (as of 22 Nov 2025) Current update level: 12 Nov 2025 (what's new) Possible actions Do nothing; all files are up to date.

-------------------------------------------
Richard Williams
Professor Emeritus of Sociology
University of Notre Dame
StataNow Version: 19.5 MP (2 processor)
EMAIL: [email protected]
WWW: https://academicweb.nd.edu/~rwilliam/
Comment

Carlo Lazzaro

Join Date: Apr 2014
Posts: 17859

22 Nov 2025, 07:06

Dear Richard,
same update date!

Code:

. update query
(contacting https://www.stata.com)

Update status
    Last check for updates:     22 Nov 2025
    New update available:       none         (as of 22 Nov 2025)
    Current update level:       12 Nov 2025  (what's new)

Possible actions

    Do nothing; all files are up to date.

.

Kind regards,
Carlo
(Stata 19.0)

Comment

George Ford

Join Date: Aug 2014

Posts: 3338
#7

22 Nov 2025, 07:51

My 19.5 was updated and it failed to produced r(table).
Comment

Carlo Lazzaro

Join Date: Apr 2014
Posts: 17859

22 Nov 2025, 08:00

I've just given it another shot (local time: UTC+1.00):

Code:

. display "`c(current_time)' `c(current_date)'"
15:57:20 22 Nov 2025

. use "C:\Program Files\Stata19\ado\base\a\auto.dta"
(1978 automobile data)

. reg price mpg, absorb(rep78)

Linear regression, absorbing indicators         Number of obs     =         69
                                                F(1, 63)          =      20.72
                                                Prob > F          =     0.0000
                                                R-squared         =     0.2584
                                                Adj R-squared     =     0.1995
                                                Root MSE          =     2605.8

------------------------------------------------------------------------------
       price | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
         mpg |  -280.2615   61.57666    -4.55   0.000    -403.3126   -157.2103
       _cons |   12112.77   1347.968     8.99   0.000      9419.07    14806.47
------------------------------------------------------------------------------

. return list

matrices:
              r(table) :  9 x 2

. mat list r(table)

r(table)[9,2]
               mpg       _cons
     b  -280.26149    12112.77
    se   61.576661   1347.9683
     t  -4.5514239   8.9859455
pvalue    .0000249   6.848e-13
    ll  -403.31262   9419.0701
    ul  -157.21035    14806.47
    df          63          63
  crit   1.9983405   1.9983405
 eform           0           0

.

Kind regards,
Carlo
(Stata 19.0)

Comment

Clyde Schechter

Join Date: Apr 2014
Posts: 30365

22 Nov 2025, 10:56

My 19.5 is also updated, I run Windows 10, and it reproduces the problem.

Code:

. update query
(contacting https://www.stata.com)

Update status
    Last check for updates:     22 Nov 2025
    New update available:       none         (as of 22 Nov 2025)
    Current update level:       12 Nov 2025  (what's new)

Possible actions

    Do nothing; all files are up to date.

. sysuse auto, clear
(1978 automobile data)

. reg price mpg, absorb(rep78)

Linear regression, absorbing indicators         Number of obs     =         69
                                                F(1, 63)          =      20.72
                                                Prob > F          =     0.0000
                                                R-squared         =     0.2584
                                                Adj R-squared     =     0.1995
                                                Root MSE          =     2605.8

------------------------------------------------------------------------------
       price | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
         mpg |  -280.2615   61.57666    -4.55   0.000    -403.3126   -157.2103
       _cons |   12112.77   1347.968     8.99   0.000      9419.07    14806.47
------------------------------------------------------------------------------

. return list

macros:
      r(datasignature) : "69:3:1885641812:1243663274"

Comment

Rich Goldstein

Join Date: Mar 2014
Posts: 4559

#10

22 Nov 2025, 11:54

as above, my version 19.5 is also up-to-date and I also do not see r(table) in this situation:

Code:

. update query
(contacting https://www.stata.com)

Update status
    Last check for updates:     22 Nov 2025
    New update available:       none         (as of 22 Nov 2025)
    Current update level:       12 Nov 2025  (what's new)

Possible actions

    Do nothing; all files are up to date.
r; t=1.00 13:51:00

. sysuse auto
(1978 automobile data)
r; t=0.00 13:51:15

. reg price mpg, absorb(rep78)

Linear regression, absorbing indicators         Number of obs     =         69
                                                F(1, 63)          =      20.72
                                                Prob > F          =     0.0000
                                                R-squared         =     0.2584
                                                Adj R-squared     =     0.1995
                                                Root MSE          =     2605.8

------------------------------------------------------------------------------
       price | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
         mpg |  -280.2615   61.57666    -4.55   0.000    -403.3126   -157.2103
       _cons |   12112.77   1347.968     8.99   0.000      9419.07    14806.47
------------------------------------------------------------------------------
r; t=0.10 13:51:26

. return li

macros:
      r(datasignature) : "69:3:1885641812:1243663274"

I am using a Mac (Tahoe 26.1)

Comment

Richard Williams

Join Date: Apr 2014

Posts: 5044
#11

22 Nov 2025, 14:46

I wonder why Carlo rates. Carlo, do you use Windows or something else?

-------------------------------------------
Richard Williams
Professor Emeritus of Sociology
University of Notre Dame
StataNow Version: 19.5 MP (2 processor)
EMAIL: [email protected]
WWW: https://academicweb.nd.edu/~rwilliam/
Comment

Clyde Schechter

Join Date: Apr 2014
Posts: 30365

#12

22 Nov 2025, 17:31

FWIW, in version 17, running on Windows, r(table) is returned:

Code:

. sysuse auto
(1978 automobile data)

. regress price mpg, absorb(rep78)

Linear regression, absorbing indicators         Number of obs     =         69
                                                F(1, 63)          =      20.72
                                                Prob > F          =     0.0000
                                                R-squared         =     0.2584
                                                Adj R-squared     =     0.1995
                                                Root MSE          =     2605.8

------------------------------------------------------------------------------
       price | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
         mpg |  -280.2615   61.57666    -4.55   0.000    -403.3126   -157.2103
       _cons |   12112.77   1347.968     8.99   0.000      9419.07    14806.47
------------------------------------------------------------------------------

. return list

matrices:
              r(table) :  9 x 2

. matrix list r(table)

r(table)[9,2]
               mpg       _cons
     b  -280.26149    12112.77
    se   61.576661   1347.9683
     t  -4.5514239   8.9859455
pvalue    .0000249   6.848e-13
    ll  -403.31262   9419.0701
    ul  -157.21035    14806.47
    df          63          63
  crit   1.9983405   1.9983405
 eform           0           0

Comment

Carlo Lazzaro

Join Date: Apr 2014

Posts: 17859
#13

23 Nov 2025, 03:05

Richard:
I use Windows 11 Home.

Kind regards,
Carlo
(Stata 19.0)
Comment
Jeff Pitblado (StataCorp)

StataCorp Employee

Join Date: Mar 2014

Posts: 755
#14

24 Nov 2025, 09:19

Unlike xtreg, fe and areg, regress does not officially support option absorb(). Unfortunately, the specified absorb() option is being passed thru to some code that is common between these commands, but regress does not handle the full range of absorb() features and responsibilities that are in areg and xtreg, fe.

Given this post and the July 2024 post referenced by Andrew in #3, we will disallow option absorb() in regress in the next Stata update.

Last edited by Jeff Pitblado (StataCorp); 24 Nov 2025, 09:38.
2 likes
Comment

Announcement