listcoef with ivreg2

Paris Rira

Join Date: Dec 2022

Posts: 383
#1

listcoef with ivreg2

02 Jul 2024, 17:01

Dear Profs and colleagues,

I need the standard deviation of coefficients so I use listcoef after reg var which works. However when I apply listcoef after ivreg2 it says "listcoef does not work with ivreg2". I believe that there might exist a common /way that is compatible with ivreg2. Any ideas are appreciated.

Code:

ivreg2 ln_labor_productivity_w (hi_nationality =IV ) logsize foreign_aff i.year i.sector i.region ltenur lfirmage multi lageworker share_9 share_12 share_uni ethnic1 ethnic2 ethnic3 ethnic4 ethnic5 ethnic6 ethnic7 ethnic8 , first robust

Cheers,
Paris
Tags: None

Andrew Musau

Join Date: Oct 2014
Posts: 9944

02 Jul 2024, 19:46

ivreg2 is from SSC, as you are asked to explain in FAQ Advice #12. If you just need the SDs of variables used by ivreg2, why not just use summarize?

Code:

webuse hsng2, clear
ivreg2 rent pcturban (hsngval = faminc i.region)
sum `e(depvar)' `e(instd)' `e(insts)' if e(sample), sep(0)

Res.:

Code:

. ivreg2 rent pcturban (hsngval = faminc i.region)

IV (2SLS) estimation
--------------------

Estimates efficient for homoskedasticity only
Statistics consistent for homoskedasticity only

                                                      Number of obs =       50
                                                      F(  2,    47) =    42.66
                                                      Prob > F      =   0.0000
Total (centered) SS     =     61243.12                Centered R2   =   0.5989
Total (uncentered) SS   =      2816856                Uncentered R2 =   0.9913
Residual SS             =  24565.71669                Root MSE      =    22.17

------------------------------------------------------------------------------
        rent | Coefficient  Std. err.      z    P>|z|     [95% conf. interval]
-------------+----------------------------------------------------------------
     hsngval |   .0022398   .0003284     6.82   0.000     .0015961    .0028836
    pcturban |    .081516   .2987652     0.27   0.785     -.504053     .667085
       _cons |   120.7065   15.22839     7.93   0.000     90.85942    150.5536
------------------------------------------------------------------------------
Underidentification test (Anderson canon. corr. LM statistic):          27.364
                                                   Chi-sq(4) P-val =    0.0000
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic):               13.298
Stock-Yogo weak ID test critical values:  5% maximal IV relative bias    16.85
                                         10% maximal IV relative bias    10.27
                                         20% maximal IV relative bias     6.71
                                         30% maximal IV relative bias     5.34
                                         10% maximal IV size             24.58
                                         15% maximal IV size             13.96
                                         20% maximal IV size             10.26
                                         25% maximal IV size              8.31
Source: Stock-Yogo (2005).  Reproduced by permission.
------------------------------------------------------------------------------
Sargan statistic (overidentification test of all instruments):          11.288
                                                   Chi-sq(3) P-val =    0.0103
------------------------------------------------------------------------------
Instrumented:         hsngval
Included instruments: pcturban
Excluded instruments: faminc 2.region 3.region 4.region
------------------------------------------------------------------------------

. 
. sum `e(depvar)' `e(instd)' `e(insts)' if e(sample), sep(0)

    Variable |        Obs        Mean    Std. dev.       Min        Max
-------------+---------------------------------------------------------
        rent |         50      234.76    35.35335        180        368
     hsngval |         50       48484    15770.24      31100     119400
      faminc |         50    19499.92    2617.218      14591      28395
             |
      region |
    N Cntrl  |         50         .24    .4314191          0          1
      South  |         50         .32    .4712121          0          1
       West  |         50         .26    .4430875          0          1
             |
    pcturban |         50    66.94913    14.40956   33.77319   91.29498

.

Comment

Paris Rira

Join Date: Dec 2022

Posts: 383
#3

03 Jul 2024, 07:02

Dear Andrew, Thank you for getting back to me.
I need SD of coefficients as I asked in #1, not variables.
Comment

Andrew Musau

Join Date: Oct 2014
Posts: 9944

03 Jul 2024, 07:39

There is no such thing as a standard deviation (SD) of a coefficient. If you run listcoef (part of spost13_ado from https://jslsoc.sitehost.iu.edu/stata), "SDofX" refers to the SD of the variable.

Code:

sysuse auto, clear
regress mpg disp weight
listcoef
sum disp weight

Res.:

Code:

.
. listcoef

regress (N=74): Unstandardized and standardized estimates

  Observed SD:  5.7855
  SD of error:  3.4561

-------------------------------------------------------------------------------
             |         b        t    P>|t|    bStdX    bStdY   bStdXY     SDofX
-------------+-----------------------------------------------------------------
displacement |    0.0053    0.535    0.594    0.485    0.001    0.084    91.837
      weight |   -0.0066   -5.631    0.000   -5.104   -0.001   -0.882   777.194
    constant |   40.0845   19.843    0.000        .        .        .         .
-------------------------------------------------------------------------------

.
. sum disp weight

    Variable |        Obs        Mean    Std. dev.       Min        Max
-------------+---------------------------------------------------------
displacement |         74    197.2973    91.83722         79        425
      weight |         74    3019.459    777.1936       1760       4

Comment

Paris Rira

Join Date: Dec 2022

Posts: 383
#5

03 Jul 2024, 08:33

Originally posted by Andrew Musau View Post

There is no such thing as a standard deviation (SD) of a coefficient.

Of course it does exit!

help listcoef

Code:

Title listcoef -- Listing regression coefficients with help for interpretation General syntax listcoef [varlist], [ pvalue(#) [ factor | percent | std ] constant reverse gt lt adjacent positive negative expand nolabel help ] where varlist contains variables from the regression model for which coefficients are to be listed. Overview listcoef lists the estimated coefficients for a variety of regression models.

Moreover, https://stats.oarc.ucla.edu/stata/we...tcoef-command/

bStdX . These are the regression coefficients with the x-variables (the independent variables) in standard deviations and the y-variable (the dependent variable) in its original units.
bStdY. These are the regression coefficients with the x-variables (the independent variables) in original units and the y-variable (the dependent variable) in standard deviations.
bStdXY . These are the regression coefficients with both the x-variable (the independent variable) and the y-variable (the dependent variable) in standard deviations. You will notice that these are the same values given in the Beta column of the regression output.
SDofX. This is the standard deviation of the x-variables (the dependent variables). I need bStdX
Comment
Andrew Musau

Join Date: Oct 2014

Posts: 9944
#6

03 Jul 2024, 10:30

Our terminologies differ. You want standardized coefficients and not standard deviations. As listcoef does not support ivreg2, you need to standardize the variables first before running ivreg2. Here is an example that shows you how to do this: #5 https://www.statalist.org/forums/for...s-for-coefplot. If you have missing values, make sure to standardize using the estimation sample.

Code:

ivreg2 ... foreach var in ... { qui sum `var' if e(sample) replace `var'= (`var'- r(mean))/r(sd) if e(sample) }

Last edited by Andrew Musau; 03 Jul 2024, 10:32.
Comment

Paris Rira

Join Date: Dec 2022
Posts: 383

03 Jul 2024, 10:53

I have to repeat my question: I would like to obtain coefficients in standard deviations which have been shown in numerous papers with exactly in this term. For not to mislead other users, especially freshmores, please use the term correctly.

The above code does not give the same result of listcoef

Code:

. reg ln_labor_productivity_w  hi_nationality lfirmage logsize foreign_aff multi lageworker ltenur  share_9 share_12 share_uni  ethn
> ic1 ethnic2  ethnic3 ethnic4 ethnic5 ethnic6 ethnic7 ethnic8 i.year i.sector i.region,  robust

Linear regression                               Number of obs     =  1,461,700
                                                F(40, 1461659)    =    6497.78
                                                Prob > F          =     0.0000
                                                R-squared         =     0.1623
                                                Root MSE          =      .8202

--------------------------------------------------------------------------------
               |               Robust
ln_labor_pro~w |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------+----------------------------------------------------------------
hi_nationality |  -.0692897   .0027147   -25.52   0.000    -.0746104    -.063969
      lfirmage |   .0561689   .0007828    71.76   0.000     .0546347    .0577031
       logsize |   .1640822   .0007768   211.21   0.000     .1625596    .1656048
   foreign_aff |   .5429106   .0055519    97.79   0.000     .5320289    .5537922
         multi |  -.0580812    .002438   -23.82   0.000    -.0628595   -.0533028
    lageworker |  -.0331365   .0031111   -10.65   0.000    -.0392341   -.0270388
        ltenur |   .0872727   .0028695    30.41   0.000     .0816486    .0928968
       share_9 |   .1841017   .0035656    51.63   0.000     .1771132    .1910901
      share_12 |   .4264239   .0039823   107.08   0.000     .4186187    .4342291
     share_uni |   .7683857   .0048659   157.91   0.000     .7588488    .7779227
       ethnic1 |  -.5000198    .035822   -13.96   0.000    -.5702297   -.4298099
       ethnic2 |  -.3399935   .0159652   -21.30   0.000    -.3712847   -.3087022
       ethnic3 |  -.1982899   .0239308    -8.29   0.000    -.2451934   -.1513865
       ethnic4 |  -.0865953   .0119784    -7.23   0.000    -.1100727    -.063118
       ethnic5 |   .1008093    .015535     6.49   0.000     .0703612    .1312575
       ethnic6 |  -.1584543   .0163181    -9.71   0.000    -.1904372   -.1264715
       ethnic7 |  -.0187402   .0794657    -0.24   0.814    -.1744902    .1370098
       ethnic8 |  -.2046155   .0103492   -19.77   0.000    -.2248996   -.1843314
               |
          year |
         2011  |  -.0735184   .0031619   -23.25   0.000    -.0797156   -.0673213
         2012  |  -.0729002   .0032489   -22.44   0.000     -.079268   -.0665324
         2013  |  -.0226434   .0032608    -6.94   0.000    -.0290344   -.0162524
         2014  |  -.0256895   .0032558    -7.89   0.000    -.0320708   -.0193083
         2015  |  -.0179962   .0031791    -5.66   0.000    -.0242271   -.0117652
         2016  |   -.010204    .003135    -3.25   0.001    -.0163485   -.0040594
         2017  |  -.0211848   .0030926    -6.85   0.000    -.0272462   -.0151234
         2018  |  -.0464156   .0030773   -15.08   0.000     -.052447   -.0403842
         2019  |  -.0687282     .00308   -22.31   0.000    -.0747649   -.0626914
               |
        sector |
            6  |   .3896998   .0022505   173.16   0.000      .385289    .3941106
            7  |   .1147885   .0019701    58.27   0.000     .1109272    .1186498
            9  |  -.4049533   .0025436  -159.21   0.000    -.4099387    -.399968
           10  |  -.0584985   .0056674   -10.32   0.000    -.0696065   -.0473906
           11  |   .2246324   .0058465    38.42   0.000     .2131735    .2360913
           12  |   .0390015   .0031416    12.41   0.000     .0328441     .045159
           13  |   .0237765    .004052     5.87   0.000     .0158348    .0317181
               |
        region |
            2  |   .0752821   .0017302    43.51   0.000      .071891    .0786731
            3  |   .1317674   .0019476    67.66   0.000     .1279501    .1355846
            4  |   .0602823   .0035086    17.18   0.000     .0534056     .067159
            5  |   .0584914   .0031386    18.64   0.000     .0523397     .064643
            6  |   .0951132   .0056546    16.82   0.000     .0840303    .1061961
            7  |  -.0051662   .0053328    -0.97   0.333    -.0156184    .0052859
               |
         _cons |   8.733179   .0117687   742.07   0.000     8.710113    8.756246
--------------------------------------------------------------------------------

.
end of do-file

. do "C:\Users\CeBER\AppData\Local\Temp\STD24c0_000000.tmp"

. listcoef hi_nationality lfirmage logsize foreign_aff multi lageworker ltenur  share_9 share_12 share_uni  ethnic1 ethnic2  ethnic3
>  ethnic4 ethnic5 ethnic6 ethnic7 ethnic8

regress (N=1461700): Unstandardized and standardized estimates

  Observed SD:  0.8961
  SD of error:  0.8202

-------------------------------------------------------------------------------
             |         b        t    P>|t|    bStdX    bStdY   bStdXY     SDofX
-------------+-----------------------------------------------------------------
hi_nationa~y |   -0.0693  -25.524    0.000   -0.019   -0.077   -0.022     0.279
    lfirmage |    0.0562   71.758    0.000    0.058    0.063    0.064     1.024
     logsize |    0.1641  211.215    0.000    0.180    0.183    0.201     1.099
 foreign_aff |    0.5429   97.787    0.000    0.079    0.606    0.088     0.145
       multi |   -0.0581  -23.824    0.000   -0.015   -0.065   -0.017     0.265
  lageworker |   -0.0331  -10.651    0.000   -0.008   -0.037   -0.009     0.255
      ltenur |    0.0873   30.414    0.000    0.033    0.097    0.037     0.375
     share_9 |    0.1841   51.633    0.000    0.069    0.205    0.077     0.377
    share_12 |    0.4264  107.080    0.000    0.140    0.476    0.156     0.329
   share_uni |    0.7684  157.913    0.000    0.231    0.857    0.257     0.300
     ethnic1 |   -0.5000  -13.958    0.000   -0.012   -0.558   -0.014     0.025
     ethnic2 |   -0.3400  -21.296    0.000   -0.017   -0.379   -0.019     0.049
     ethnic3 |   -0.1983   -8.286    0.000   -0.007   -0.221   -0.007     0.034
     ethnic4 |   -0.0866   -7.229    0.000   -0.006   -0.097   -0.007     0.069
     ethnic5 |    0.1008    6.489    0.000    0.006    0.112    0.007     0.059
     ethnic6 |   -0.1585   -9.710    0.000   -0.008   -0.177   -0.009     0.053
     ethnic7 |   -0.0187   -0.236    0.814   -0.000   -0.021   -0.000     0.012
     ethnic8 |   -0.2046  -19.771    0.000   -0.017   -0.228   -0.019     0.084
-------------------------------------------------------------------------------

So based on the above code the coefficient of hi_natioanlity must be -0.019

Code:

. foreach var in  ln_labor_productivity_w  hi_nationality lfirmage logsize foreign_aff multi lageworker ltenur  share_9 share_12 sha
> re_uni  ethnic1 ethnic2  ethnic3 ethnic4 ethnic5 ethnic6 ethnic7 ethnic8 {
  2.     qui sum `var'
  3.     replace `var'= (`var'- r(mean))/r(sd)
  4. }
(2,309,140 real changes made)
(1,544,719 real changes made)
(1,461,700 real changes made)
(1,544,719 real changes made)
(2,309,140 real changes made)
(1,544,719 real changes made)
(1,544,719 real changes made)
(1,544,719 real changes made)
(1,544,719 real changes made)
(1,544,719 real changes made)
(1,544,719 real changes made)
(1,544,719 real changes made)
(1,544,719 real changes made)
(1,544,719 real changes made)
(1,544,719 real changes made)
(1,544,719 real changes made)
(1,544,719 real changes made)
(1,544,719 real changes made)
(1,544,719 real changes made)

. regress ln_labor_productivity_w  hi_nationality lfirmage logsize foreign_aff multi lageworker ltenur  share_9 share_12 share_uni  
> ethnic1 ethnic2  ethnic3 ethnic4 ethnic5 ethnic6 ethnic7 ethnic8 i.year i.sector i.region,  robust

Linear regression                               Number of obs     =  1,461,700
                                                F(40, 1461659)    =    6497.78
                                                Prob > F          =     0.0000
                                                R-squared         =     0.1623
                                                Root MSE          =     .72388

--------------------------------------------------------------------------------
               |               Robust
ln_labor_pro~w |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------+----------------------------------------------------------------
hi_nationality |  -.0170773   .0006691   -25.52   0.000    -.0183887   -.0157659
      lfirmage |   .0507489   .0007072    71.76   0.000     .0493628     .052135
       logsize |   .1588333    .000752   211.21   0.000     .1573594    .1603072
   foreign_aff |   .0647823   .0006625    97.79   0.000     .0634839    .0660808
         multi |  -.0134323   .0005638   -23.82   0.000    -.0145374   -.0123272
    lageworker |  -.0075215   .0007062   -10.65   0.000    -.0089056   -.0061374
        ltenur |   .0295718   .0009723    30.41   0.000     .0276661    .0314775
       share_9 |   .0614998   .0011911    51.63   0.000     .0591653    .0638343
      share_12 |   .1246938   .0011645   107.08   0.000     .1224114    .1269761
     share_uni |   .2044285   .0012946   157.91   0.000     .2018912    .2069658
       ethnic1 |  -.0114241   .0008184   -13.96   0.000    -.0130283     -.00982
       ethnic2 |  -.0152237   .0007149   -21.30   0.000    -.0166248   -.0138226
       ethnic3 |  -.0060353   .0007284    -8.29   0.000    -.0074629   -.0046077
       ethnic4 |  -.0054209   .0007498    -7.23   0.000    -.0068905   -.0039512
       ethnic5 |   .0053514   .0008247     6.49   0.000      .003735    .0069677
       ethnic6 |  -.0076401   .0007868    -9.71   0.000    -.0091822    -.006098
       ethnic7 |  -.0001969    .000835    -0.24   0.814    -.0018335    .0014397
       ethnic8 |  -.0158865   .0008035   -19.77   0.000    -.0174614   -.0143117
               |
          year |
         2011  |  -.0648847   .0027905   -23.25   0.000    -.0703541   -.0594153
         2012  |  -.0643391   .0028674   -22.44   0.000    -.0699591   -.0587191
         2013  |  -.0199843   .0028778    -6.94   0.000    -.0256247   -.0143438
         2014  |  -.0226727   .0028734    -7.89   0.000    -.0283045   -.0170408
         2015  |  -.0158828   .0028058    -5.66   0.000     -.021382   -.0103836
         2016  |  -.0090057   .0027669    -3.25   0.001    -.0144286   -.0035827
         2017  |  -.0186969   .0027294    -6.85   0.000    -.0240465   -.0133473
         2018  |  -.0409647   .0027159   -15.08   0.000    -.0462878   -.0356416
         2019  |   -.060657   .0027183   -22.31   0.000    -.0659848   -.0553292
               |
        sector |
            6  |   .3439351   .0019862   173.16   0.000     .3400422    .3478279
            7  |   .1013082   .0017387    58.27   0.000     .0979004    .1047161
            9  |  -.3573973   .0022449  -159.21   0.000    -.3617972   -.3529975
           10  |  -.0516287   .0050019   -10.32   0.000    -.0614322   -.0418252
           11  |   .1982525   .0051599    38.42   0.000     .1881393    .2083658
           12  |   .0344214   .0027727    12.41   0.000      .028987    .0398557
           13  |   .0209843   .0035761     5.87   0.000     .0139752    .0279933
               |
        region |
            2  |   .0664413    .001527    43.51   0.000     .0634484    .0694341
            3  |   .1162932   .0017189    67.66   0.000     .1129242    .1196621
            4  |    .053203   .0030965    17.18   0.000     .0471339    .0592721
            5  |   .0516224   .0027701    18.64   0.000     .0461932    .0570516
            6  |   .0839435   .0049906    16.82   0.000     .0741621    .0937249
            7  |  -.0045595   .0047066    -0.97   0.333    -.0137843    .0046652
               |
         _cons |     .08606   .0025113    34.27   0.000     .0811379    .0909821
--------------------------------------------------------------------------------

But it is -.017

Comment

Andrew Musau

Join Date: Oct 2014

Posts: 9944
#8

03 Jul 2024, 12:15

I would agree to "coefficients in standard deviations", implying that their units are standard deviations. But they are still coefficients. We also call them standardized coefficients. On your question, see my comment in #6 with regards to using the estimation sample to standardize. Post again if the results do not match.
Comment

Paris Rira

Join Date: Dec 2022
Posts: 383

03 Jul 2024, 13:01

Still, do not match.

Code:

 foreach var in ln_labor_productivity_w hi_nationality lfirmage logsize foreign_aff multi lageworker ltenur  share_9 share_12 share
> _uni ethnic1 ethnic2  ethnic3 ethnic4 ethnic5 ethnic6 ethnic7 ethnic8  {
  2.     qui sum `var' if e(sample)
  3.     replace `var'= (`var'- r(mean))/r(sd) if e(sample)
  4. }
(1,461,700 real changes made)
(1,461,700 real changes made)
(1,461,700 real changes made)
(1,461,700 real changes made)
(1,461,700 real changes made)
(1,461,700 real changes made)
(1,461,700 real changes made)
(1,461,700 real changes made)
(1,461,700 real changes made)
(1,461,700 real changes made)
(1,461,700 real changes made)
(1,461,700 real changes made)
(1,461,700 real changes made)
(1,461,700 real changes made)
(1,461,700 real changes made)
(1,461,700 real changes made)
(1,461,700 real changes made)
(1,461,700 real changes made)
(1,461,700 real changes made)

. regress ln_labor_productivity_w  hi_nationality lfirmage logsize foreign_aff multi lageworker ltenur  share_9 share_12 share_uni  
> ethnic1 ethnic2  ethnic3 ethnic4 ethnic5 ethnic6 ethnic7 ethnic8 i.year i.sector i.region,  robust

Linear regression                               Number of obs     =  1,461,700
                                                F(40, 1461659)    =    6497.78
                                                Prob > F          =     0.0000
                                                R-squared         =     0.1623
                                                Root MSE          =     .91528

--------------------------------------------------------------------------------
               |               Robust
ln_labor_pro~w |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------+----------------------------------------------------------------
hi_nationality |  -.0215514   .0008444   -25.52   0.000    -.0232064   -.0198965
      lfirmage |   .0641674   .0008942    71.76   0.000     .0624148      .06592
       logsize |   .2011429   .0009523   211.21   0.000     .1992764    .2030094
   foreign_aff |   .0880744   .0009007    97.79   0.000     .0863091    .0898397
         multi |  -.0171896   .0007215   -23.82   0.000    -.0186038   -.0157755
    lageworker |  -.0094346   .0008858   -10.65   0.000    -.0111707   -.0076984
        ltenur |   .0365032   .0012002    30.41   0.000     .0341508    .0388555
       share_9 |   .0773618   .0014983    51.63   0.000     .0744252    .0802984
      share_12 |   .1564179   .0014608   107.08   0.000     .1535549    .1592809
     share_uni |   .2573769   .0016299   157.91   0.000     .2541824    .2605714
       ethnic1 |  -.0138556   .0009926   -13.96   0.000    -.0158011   -.0119101
       ethnic2 |  -.0186202   .0008744   -21.30   0.000    -.0203339   -.0169064
       ethnic3 |  -.0074314   .0008969    -8.29   0.000    -.0091892   -.0056735
       ethnic4 |   -.006686   .0009249    -7.23   0.000    -.0084987   -.0048733
       ethnic5 |   .0065945   .0010162     6.49   0.000     .0046027    .0085863
       ethnic6 |  -.0094248   .0009706    -9.71   0.000    -.0113271   -.0075225
       ethnic7 |  -.0002413   .0010234    -0.24   0.814    -.0022471    .0017644
       ethnic8 |  -.0191722   .0009697   -19.77   0.000    -.0210728   -.0172716
               |
          year |
         2011  |  -.0820409   .0035284   -23.25   0.000    -.0889564   -.0751253
         2012  |   -.081351   .0036256   -22.44   0.000     -.088457    -.074245
         2013  |  -.0252683   .0036388    -6.94   0.000    -.0324002   -.0181365
         2014  |  -.0286675   .0036332    -7.89   0.000    -.0357885   -.0215466
         2015  |  -.0200823   .0035476    -5.66   0.000    -.0270356   -.0131291
         2016  |  -.0113868   .0034984    -3.25   0.001    -.0182436     -.00453
         2017  |  -.0236406   .0034511    -6.85   0.000    -.0304047   -.0168765
         2018  |  -.0517962    .003434   -15.08   0.000    -.0585268   -.0450656
         2019  |  -.0766953   .0034371   -22.31   0.000    -.0834319   -.0699588
               |
        sector |
            6  |    .434875   .0025113   173.16   0.000     .4299528    .4397971
            7  |   .1280952   .0021985    58.27   0.000     .1237862    .1324041
            9  |  -.4518968   .0028384  -159.21   0.000      -.45746   -.4463335
           10  |  -.0652799   .0063244   -10.32   0.000    -.0776755   -.0528842
           11  |   .2506725   .0065242    38.42   0.000     .2378852    .2634598
           12  |   .0435227   .0035058    12.41   0.000     .0366515    .0503939
           13  |   .0265327   .0045217     5.87   0.000     .0176704     .035395
               |
        region |
            2  |    .084009   .0019307    43.51   0.000     .0802248    .0877932
            3  |   .1470422   .0021734    67.66   0.000     .1427824     .151302
            4  |   .0672704   .0039153    17.18   0.000     .0595966    .0749443
            5  |   .0652719   .0035025    18.64   0.000     .0584071    .0721366
            6  |    .106139   .0063101    16.82   0.000     .0937713    .1185067
            7  |  -.0057651    .005951    -0.97   0.333     -.017429    .0058987
               |
         _cons |  -.0879333   .0031711   -27.73   0.000    -.0941485    -.081718
--------------------------------------------------------------------------------

Comment

Andrew Musau

Join Date: Oct 2014

Posts: 9944
#10

03 Jul 2024, 13:16

Reduce the number of variables and observations and find a subsample that reproduces the issue. Then provide a dataex example of this subsample.
Comment

Paris Rira

Join Date: Dec 2022
Posts: 383

#11

03 Jul 2024, 14:22

Code:

clear
input float(ln_labor_productivity_w hi_nationality)
 9.281988 .9923468
 9.333387 .9915874
  8.76686 .9959396
10.198627 .9926839
 9.510689 .9909202
 8.791356   .99805
10.330937 .9905199
 9.590218 .9978104
10.110822 .9940464
 9.637493 .9989997
  9.69247  .995016
 9.546683 .9903305
 8.837241 .9992133
  8.22231 .9952798
 9.584678 .9944996
 8.645433 .9975001
 9.868996 .9953218
10.096255 .9922356
 9.187426 .9972121
 8.821714 .9942925
  10.3666 .9971483
 9.302102 .9982308
 8.239303 .9930201
 9.406878 .9906753
 9.532752  .994872
 9.996658 .9967926
 8.682385 .9977748
 8.732817 .9961808
 8.695457 .9961513
 8.513886 .9924938
10.130157 .9904519
 9.382533 .9930528
 8.517624 .9947981
10.396965 .9931585
8.9895525 .9984176
 8.781916 .9976585
 8.876676  .998732
 9.336425 .9915495
 9.252281  .998728
10.703698 .9911469
 9.224909 .9938857
 9.494924 .9900637
 9.044812 .9986701
 9.090109 .9957239
 9.526831 .9927593
 8.741158 .9900004
 8.975822 .9976693
 9.646744 .9943078
 9.582047 .9962072
 8.703842 .9966113
  9.44824 .9953623
 9.600907 .9952037
 8.733271 .9964204
 9.358811 .9968798
9.1950245 .9913253
 8.698541 .9943939
9.1031475 .9957154
 8.942465 .9985592
 8.422465 .9935536
 9.357276 .9985983
 9.093091 .9911243
 8.673805 .9924456
  10.7588 .9911495
  8.94105 .9969143
 9.669009 .9936522
 8.645086 .9960824
 9.706211 .9911944
 8.966554 .9913761
 9.657453 .9925055
 9.553917 .9961549
 9.505138 .9934007
 9.144127 .9939017
  8.71483  .990463
 8.838542 .9961025
10.076946 .9914981
 9.047061 .9941899
 9.340981  .990203
 9.635928 .9945736
 8.226807 .9903846
 9.242423 .9924413
end

It must be -0.143 though it is -.252

Code:

 reg ln_labor_productivity_w  hi_nationality

      Source |       SS           df       MS      Number of obs   =        80
-------------+----------------------------------   F(1, 78)        =      5.33
       Model |  1.61857633         1  1.61857633   Prob > F        =    0.0236
    Residual |  23.6964161        78  .303800206   R-squared       =    0.0639
-------------+----------------------------------   Adj R-squared   =    0.0519
       Total |  25.3149924        79  .320442942   Root MSE        =    .55118

--------------------------------------------------------------------------------
ln_labor_pro~w |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------+----------------------------------------------------------------
hi_nationality |  -52.50897   22.74893    -2.31   0.024    -97.79861   -7.219333
         _cons |   61.49565    22.6229     2.72   0.008     16.45691    106.5344
--------------------------------------------------------------------------------

. listcoef hi_nationality

regress (N=80): Unstandardized and standardized estimates 

  Observed SD:  0.5661
  SD of error:  0.5512

-------------------------------------------------------------------------------
             |         b        t    P>|t|    bStdX    bStdY   bStdXY     SDofX
-------------+-----------------------------------------------------------------
hi_nationa~y |  -52.5090   -2.308    0.024   -0.143  -92.759   -0.253     0.003
-------------------------------------------------------------------------------

. preserve
already preserved
r(621);

. foreach var in ln_labor_productivity_w hi_nationality  {
  2. 
.     qui sum `var' if e(sample)
  3. 
.     replace `var'= (`var'- r(mean))/r(sd) if e(sample)
  4. 
. }
(80 real changes made)
(80 real changes made)

. reg ln_labor_productivity_w  hi_nationality

      Source |       SS           df       MS      Number of obs   =        80
-------------+----------------------------------   F(1, 78)        =      5.33
       Model |  5.05105959         1  5.05105959   Prob > F        =    0.0236
    Residual |  73.9489399        78  .948063332   R-squared       =    0.0639
-------------+----------------------------------   Adj R-squared   =    0.0519
       Total |  78.9999995        79  .999999993   Root MSE        =    .97369

--------------------------------------------------------------------------------
ln_labor_pro~w |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
---------------+----------------------------------------------------------------
hi_nationality |  -.2528586   .1095482    -2.31   0.024    -.4709522   -.0347649
         _cons |   4.69e-09   .1088613     0.00   1.000    -.2167263    .2167263
--------------------------------------------------------------------------------

Comment

Andrew Musau

Join Date: Oct 2014
Posts: 9944

#12

03 Jul 2024, 15:32

"X" in bStd[X|Y|XY] refers to the vector of RHS variables and "Y" is the outcome. Indeed, your standardized coefficient does match "bStdXY". For what you want, just standardize the RHS variables.

Code:

clear
input float(ln_labor_productivity_w hi_nationality)
 9.281988 .9923468
 9.333387 .9915874
  8.76686 .9959396
10.198627 .9926839
 9.510689 .9909202
 8.791356   .99805
10.330937 .9905199
 9.590218 .9978104
10.110822 .9940464
 9.637493 .9989997
  9.69247  .995016
 9.546683 .9903305
 8.837241 .9992133
  8.22231 .9952798
 9.584678 .9944996
 8.645433 .9975001
 9.868996 .9953218
10.096255 .9922356
 9.187426 .9972121
 8.821714 .9942925
  10.3666 .9971483
 9.302102 .9982308
 8.239303 .9930201
 9.406878 .9906753
 9.532752  .994872
 9.996658 .9967926
 8.682385 .9977748
 8.732817 .9961808
 8.695457 .9961513
 8.513886 .9924938
10.130157 .9904519
 9.382533 .9930528
 8.517624 .9947981
10.396965 .9931585
8.9895525 .9984176
 8.781916 .9976585
 8.876676  .998732
 9.336425 .9915495
 9.252281  .998728
10.703698 .9911469
 9.224909 .9938857
 9.494924 .9900637
 9.044812 .9986701
 9.090109 .9957239
 9.526831 .9927593
 8.741158 .9900004
 8.975822 .9976693
 9.646744 .9943078
 9.582047 .9962072
 8.703842 .9966113
  9.44824 .9953623
 9.600907 .9952037
 8.733271 .9964204
 9.358811 .9968798
9.1950245 .9913253
 8.698541 .9943939
9.1031475 .9957154
 8.942465 .9985592
 8.422465 .9935536
 9.357276 .9985983
 9.093091 .9911243
 8.673805 .9924456
  10.7588 .9911495
  8.94105 .9969143
 9.669009 .9936522
 8.645086 .9960824
 9.706211 .9911944
 8.966554 .9913761
 9.657453 .9925055
 9.553917 .9961549
 9.505138 .9934007
 9.144127 .9939017
  8.71483  .990463
 8.838542 .9961025
10.076946 .9914981
 9.047061 .9941899
 9.340981  .990203
 9.635928 .9945736
 8.226807 .9903846
 9.242423 .9924413
end

qui reg ln_labor_productivity_w  hi_nationality
foreach var in  hi_nationality  {
  qui sum `var' if e(sample)
  replace `var'= (`var'- r(mean))/r(sd) if e(sample)
}
reg ln_labor_productivity_w  hi_nationality

Res.:

Code:

. qui reg ln_labor_productivity_w  hi_nationality

.
. foreach var in  hi_nationality  {
  2.
.   qui sum `var' if e(sample)
  3.
.   replace `var'= (`var'- r(mean))/r(sd) if e(sample)
  4.
. }
(80 real changes made)

.
. reg ln_labor_productivity_w  hi_nationality

      Source |       SS           df       MS      Number of obs   =        80
-------------+----------------------------------   F(1, 78)        =      5.33
       Model |  1.61859692         1  1.61859692   Prob > F        =    0.0236
    Residual |  23.6963981        78  .303799976   R-squared       =    0.0639
-------------+----------------------------------   Adj R-squared   =    0.0519
       Total |  25.3149951        79  .320442976   Root MSE        =    .55118

--------------------------------------------------------------------------------
ln_labor_pro~w | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
---------------+----------------------------------------------------------------
hi_nationality |  -.1431383   .0620127    -2.31   0.024     -.266596   -.0196806
         _cons |   9.277767   .0616239   150.55   0.000     9.155083    9.400451
--------------------------------------------------------------------------------

Comment

Paris Rira

Join Date: Dec 2022

Posts: 383
#13

03 Jul 2024, 16:09

That's correct! Only RHS works. Thank you so much.
Comment

Announcement