Announcement

Collapse
No announcement yet.
X
Collapse
  • Page of 1
  • Filter
  • Time
  • Show
Clear All
new posts
  • #1

    Large Negative Inverse Mills Ratio in Heckman Model

    Hi all,

    I'm estimating a Heckman two-step model to assess the impact of pension freezes on firms’ environmental disclosure scores. In the second stage, the inverse Mills ratio has a large, negative, and significant coefficient (–133.39, p = 0.013).

    Key variables:
    • hard_final_Exact_new: = 1 in the year a firm freezes its defined benefit (DB) pension plan for all employees, 0 otherwise.
    • csopresence1: = 1 if the firm has a Chief Sustainability Officer (CSO) in a given year, 0 otherwise.
    • FREEZEXcso: interaction between the two variables.
    The model is identified using exclusion restrictions included only in the first stage.

    Does such a large negative coefficient on the inverse Mills ratio seem plausible? Is this a problem, or could it reasonably suggest that unobserved factors associated with pension freezes are strongly linked to lower environmental disclosure?

    Any insights on why the IMR effect might be so strong — and whether this should raise concerns — would be greatly appreciated.

    these are the codes i used


    clear all
    cd "C:\Users\lenovo\Desktop\Data\source\DATA\work \thi rd analysis"
    use "anothertry"


    gen avarage = ( ENVIRON_DISCLOSURE_SCORE + SOCIAL_DISCLOSURE_SCORE ) / 2

    winsor2 ENVIRON_DISCLOSURE_SCORE SOCIAL_DISCLOSURE_SCORE avarage , cut(1 99)

    gen FREEZEXcso = hard_final_Exact_new * csopresence1


    asdoc probit hard_final_Exact_new Firm_Size_w ROA_w Leverage_w Market_book_four_w Non_pension_CFO_w STD_CFO_w board_size_w GenderRatiogenderratio_w independent_percentage_w Fund_Status_w FUNDING_RATIO_w Platn_Size_w i.year i.ff_12 , robust cluster (id) nest replace drop(i.year i.ff_12 ) dec(4) save(qqqq)

    predict lefthat, xb // Get the linear prediction
    gen mills = normalden(lefthat) / normal(lefthat) // Generate the inverse Mill's ratio


    asdoc reg ENVIRON_DISCLOSURE_SCORE_w hard_final_Exact_new csopresence1 FREEZEXcso Firm_Size_w ROA_w Leverage_w Market_book_four_w Non_pension_CFO_w STD_CFO_w board_size_w GenderRatiogenderratio_w independent_percentage_w sustainability2_w SUSTAIBILITY_COMITEE_FU Fund_Status_w FUNDING_RATIO_w Platn_Size_w mills i.year i.ff_12 , robust cluster (id) nest replace drop(i.year i.ff_12 ) dec(4) save(qqqkkq)


    Thanks in advance!











    HTML Code:
    . asdoc probit  hard_final_Exact_new   Firm_Size_w   ROA_w     Leverage_w   Market_book_four
> _w   Non_pension_CFO_w   STD_CFO_w  board_size_w  GenderRatiogenderratio_w  independent_pe
> rcentage_w    Fund_Status_w  FUNDING_RATIO_w  Platn_Size_w  i.year    i.ff_12  ,  robust c
> luster (id)  nest replace  drop(i.year i.ff_12  ) dec(4)  save(qqqq)

Iteration 0:  Log pseudolikelihood = -381.41764  
Iteration 1:  Log pseudolikelihood = -349.64094  
Iteration 2:  Log pseudolikelihood = -346.70631  
Iteration 3:  Log pseudolikelihood = -346.66482  
Iteration 4:  Log pseudolikelihood = -346.66469  
Iteration 5:  Log pseudolikelihood = -346.66469  

Probit regression                                       Number of obs =  3,854
                                                        Wald chi2(41) =  95.97
                                                        Prob > chi2   = 0.0000
Log pseudolikelihood = -346.66469                       Pseudo R2     = 0.0911

                                               (Std. err. adjusted for 282 clusters in id)
------------------------------------------------------------------------------------------
                         |               Robust
    hard_final_Exact_new | Coefficient  std. err.      z    P>|z|     [95% conf. interval]
-------------------------+----------------------------------------------------------------
             Firm_Size_w |  -.1486931   .0793764    -1.87   0.061     -.304268    .0068818
                   ROA_w |   1.394501   1.266187     1.10   0.271     -1.08718    3.876183
              Leverage_w |  -.9995838   .4151587    -2.41   0.016     -1.81328   -.1858876
      Market_book_four_w |   .0046217   .0086523     0.53   0.593    -.0123365      .02158
       Non_pension_CFO_w |  -3.994935   1.690991    -2.36   0.018    -7.309216   -.6806539
               STD_CFO_w |    3.08323   2.895561     1.06   0.287    -2.591966    8.758425
            board_size_w |  -.0078393   .0275714    -0.28   0.776    -.0618782    .0461997
GenderRatiogenderratio_w |   -.311062   .6720613    -0.46   0.643    -1.628278    1.006154
independent_percentage_w |   .4727935   .6556422     0.72   0.471    -.8122416    1.757829
           Fund_Status_w |   -3.97355   2.085994    -1.90   0.057    -8.062024    .1149231
         FUNDING_RATIO_w |  -.2610834   .4042837    -0.65   0.518    -1.053465    .5312982
            Platn_Size_w |   .0608101   .0651693     0.93   0.351    -.0669194    .1885397
                         |
                    year |
                   2005  |  -.2919346     .42794    -0.68   0.495    -1.130682    .5468125
                   2006  |  -.0162893   .3777238    -0.04   0.966    -.7566145    .7240358
                   2007  |  -.2363701   .4284376    -0.55   0.581    -1.076092    .6033522
                   2008  |    .454851   .3243855     1.40   0.161     -.180933    1.090635
                   2009  |   .6437069   .3145637     2.05   0.041     .0271733     1.26024
                   2010  |   .4648595   .3274794     1.42   0.156    -.1769883    1.106707
                   2011  |    .209522   .3635784     0.58   0.564    -.5030786    .9221227
                   2012  |   .3991531   .3388834     1.18   0.239    -.2650462    1.063352
                   2013  |   .4619265   .3408755     1.36   0.175    -.2061772     1.13003
                   2014  |   .5627522   .3245501     1.73   0.083    -.0733542    1.198859
                   2015  |   .5361479   .3424661     1.57   0.117    -.1350734    1.207369
                   2016  |   .3776218   .3464051     1.09   0.276    -.3013198    1.056563
                   2017  |   .4563485   .3514999     1.30   0.194    -.2325787    1.145276
                   2018  |   .7855175   .3368549     2.33   0.020     .1252941    1.445741
                   2019  |   .5433685   .3639084     1.49   0.135     -.169879    1.256616
                   2020  |   .8101857   .3527143     2.30   0.022     .1188783    1.501493
                   2021  |    .409564   .4195871     0.98   0.329    -.4128115     1.23194
                   2022  |   .1266738   .4616397     0.27   0.784    -.7781234    1.031471
                         |
                   ff_12 |
                      2  |   .0650237   .2984135     0.22   0.828     -.519856    .6499033
                      3  |  -.2066733   .2128754    -0.97   0.332    -.6239014    .2105547
                      4  |  -.3422767   .3916597    -0.87   0.382    -1.109916    .4253622
                      5  |  -.2263999   .2333209    -0.97   0.332    -.6837004    .2309006
                      6  |  -.0542146     .23368    -0.23   0.817    -.5122188    .4037897
                      7  |   .1257675   .4124482     0.30   0.760    -.6826161     .934151
                      8  |  -.3136073   .2436409    -1.29   0.198    -.7911347    .1639201
                      9  |   .4064531   .2035085     2.00   0.046     .0075838    .8053223
                     10  |  -.1490405   .2554473    -0.58   0.560     -.649708     .351627
                     11  |   .2807591     .22041     1.27   0.203    -.1512366    .7127549
                     12  |  -.0448456   .2166022    -0.21   0.836    -.4693781     .379687
                         |
                   _cons |  -.9685873   .9632249    -1.01   0.315    -2.856473    .9192987
------------------------------------------------------------------------------------------
Click to Open File:  qqqq.doc

. 
. predict lefthat, xb // Get the linear prediction
(4122 missing values generated)

. gen mills = normalden(lefthat) / normal(lefthat) // Generate the inverse Mill's ratio
(4,122 missing values generated)

. 
. 
. asdoc reg ENVIRON_DISCLOSURE_SCORE_w hard_final_Exact_new   csopresence1  FREEZEXcso Firm_
> Size_w   ROA_w     Leverage_w   Market_book_four_w   Non_pension_CFO_w   STD_CFO_w  board_
> size_w  GenderRatiogenderratio_w  independent_percentage_w sustainability2_w SUSTAIBILITY_
> COMITEE_FU    Fund_Status_w  FUNDING_RATIO_w  Platn_Size_w mills  i.year    i.ff_12 ,   ro
> bust cluster (id) nest replace  drop(i.year i.ff_12  ) dec(4)  save(qqqkkq)

Linear regression                               Number of obs     =      3,053
                                                F(46, 251)        =      60.40
                                                Prob > F          =     0.0000
                                                R-squared         =     0.6882
                                                Root MSE          =     12.677

                                               (Std. err. adjusted for 252 clusters in id)
------------------------------------------------------------------------------------------
                         |               Robust
ENVIRON_DISCLOSURE_SCO~w | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
-------------------------+----------------------------------------------------------------
    hard_final_Exact_new |   8.920114   3.485295     2.56   0.011     2.055963    15.78426
            csopresence1 |   6.482304   1.253054     5.17   0.000     4.014465    8.950143
              FREEZEXcso |   10.30082    5.88566     1.75   0.081     -1.29075    21.89239
             Firm_Size_w |     16.805   7.064002     2.38   0.018     2.892727    30.71727
                   ROA_w |  -188.2846   66.45055    -2.83   0.005    -319.1563    -57.4129
              Leverage_w |   122.6556   48.26031     2.54   0.012     27.60882    217.7024
      Market_book_four_w |  -.6168643    .228682    -2.70   0.007    -1.067244   -.1664841
       Non_pension_CFO_w |   522.4601   189.6096     2.76   0.006     149.0316    895.8886
               STD_CFO_w |  -349.5499    148.372    -2.36   0.019    -641.7627   -57.33712
            board_size_w |   1.290604    .436255     2.96   0.003     .4314169    2.149791
GenderRatiogenderratio_w |   35.71305   16.04088     2.23   0.027     4.121182    67.30493
independent_percentage_w |  -54.92927   23.29206    -2.36   0.019    -100.8021   -9.056474
       sustainability2_w |   .5414999   .0373389    14.50   0.000     .4679623    .6150374
 SUSTAIBILITY_COMITEE_FU |   2.366581   1.184138     2.00   0.047     .0344671    4.698694
           Fund_Status_w |   477.1368   186.8281     2.55   0.011     109.1864    845.0873
         FUNDING_RATIO_w |   28.16416   13.19009     2.14   0.034     2.186802    54.14151
            Platn_Size_w |  -6.319805   2.957543    -2.14   0.034    -12.14457   -.4950405
                   mills |  -133.3892   53.35121    -2.50   0.013    -238.4623   -28.31607
                         |
                    year |
                   2006  |  -31.80994   13.72755    -2.32   0.021     -58.8458   -4.774077
                   2007  |  -4.085754    3.58242    -1.14   0.255    -11.14119     2.96968
                   2008  |  -88.11516   36.49273    -2.41   0.016    -159.9861   -16.24418
                   2009  |  -110.2062   45.46549    -2.42   0.016    -199.7487   -20.66372
                   2010  |  -87.97598   36.99328    -2.38   0.018    -160.8328   -15.11919
                   2011  |   -56.3718   24.86244    -2.27   0.024    -105.3374    -7.40622
                   2012  |  -76.97707   33.90849    -2.27   0.024    -143.7585   -10.19564
                   2013  |  -83.63576   36.86525    -2.27   0.024    -156.2404   -11.03111
                   2014  |  -93.10664   41.60758    -2.24   0.026    -175.0511   -11.16216
                   2015  |  -86.56858   40.33304    -2.15   0.033    -166.0029   -7.134258
                   2016  |  -66.77477   32.86188    -2.03   0.043    -131.4949   -2.054605
                   2017  |  -73.28445    36.5932    -2.00   0.046    -145.3533     -1.2156
                   2018  |  -112.4621   52.15727    -2.16   0.032    -215.1837   -9.740401
                   2019  |  -82.22986   40.63829    -2.02   0.044    -162.2654   -2.194366
                   2020  |  -113.6532   53.25484    -2.13   0.034    -218.5365   -8.769923
                   2021  |  -64.27849   34.37595    -1.87   0.063    -131.9806    3.423572
                   2022  |  -29.20575   20.68805    -1.41   0.159    -69.95003    11.53854
                         |
                   ff_12 |
                      2  |  -14.28764   4.955029    -2.88   0.004    -24.04637   -4.528907
                      3  |   23.79015   10.37081     2.29   0.023     3.365261    44.21504
                      4  |   42.69021   17.53024     2.44   0.016     8.165093    77.21532
                      5  |   34.90489   11.15738     3.13   0.002     12.93088     56.8789
                      6  |   6.434335   3.873674     1.66   0.098    -1.194711    14.06338
                      7  |   -20.0249   7.356159    -2.72   0.007    -34.51256   -5.537236
                      8  |   45.84821   15.51659     2.95   0.003     15.28891    76.40751
                      9  |  -61.87624   20.05014    -3.09   0.002    -101.3642    -22.3883
                     10  |   21.75348   7.807923     2.79   0.006     6.376088    37.13087
                     11  |  -41.42383   13.41106    -3.09   0.002    -67.83638   -15.01128
                     12  |   6.663091   3.655356     1.82   0.070    -.5359882    13.86217
                         |
                   _cons |   211.2848   92.80652     2.28   0.024       28.506    394.0635
------------------------------------------------------------------------------------------
Click to Open File:  qqqkkq.doc

. 
end of do-file
    Last edited by hussein bataineh; 23 Apr 2025, 10:55.
    Tags: None
  • #2
    This post s a follow-up to your earlier post on the same topic at https://www.statalist.org/forums/for...s-this-logical. Please do not start duplicate threads.

    You have still not told us why you are trying to 'hand code' your Heckman two-stage estimator, rather than using -heckman- (with or without the two-step option). Why not use the built-in commands??

    [And by the way there's no need to use -asdoc- (SSC -- as you should tell us) when reporting output here; it's another layer of complication.]

    The output in your previous post, showing non-convergence, suggested to me collinearity among covariates.

    The Mills Ratio in Heckman's Two-Step estimator is related to the covariance between the errors in the two equations of the model. If you get a huge negative, doesn't that suggest the software is telling you the estimated correlation between errors is approaching -1? (Read a good text about this or, indeed, the Methods and Formulae section in the manual entry for -heckman-.) Which suggests something is wrong with your model specification. Check your eplanatory variables again.
    • 1 like

    Comment

    • #3
      Thank you for your feedback, and I sincerely apologize for the duplicate post—I should have followed up in the original thread instead.

      Regarding your question, I didn’t use the built-in heckman command with the twostep option initially because it wasn’t producing any results when I attempted to run it. That led me to try and implement the two-stage procedure manually.

      I now suspect the root of the issue is that both the key independent variable (csopresence1) and the moderator (FREEZEXcso) are binary, which may be causing collinearity—especially when including the interaction term. This might also explain the unusual estimates and convergence issues.

      I appreciate the guidance, and I’ll revisit the specification and diagnostics as advised. Thanks again for your time and insights.


      these are the results when i use the code
      HTML Code:
      . heckman ENVIRON_DISCLOSURE_SCORE_w csopresence1 FREEZEXcso Firm_Size_w ROA_w Leverage_w Ma
> rket_book_four_w Non_pension_CFO_w STD_CFO_w board_size_w GenderRatiogenderratio_w indepen
> dent_percentage_w sustainability2_w SUSTAIBILITY_COMITEE_FU Fund_Status_w FUNDING_RATIO_w 
> Platn_Size_w  i.year i.ff_12, select(hard_final_Exact_new = Firm_Size_w ROA_w Leverage_w M
> arket_book_four_w Non_pension_CFO_w STD_CFO board_size_w GenderRatiogenderratio_w independ
> ent_percentage_w Fund_Status_w FUNDING_RATIO_w Platn_Size_w i.year i.ff_12) twostep first
note: FREEZEXcso omitted because of collinearity.

Iteration 0:  Log likelihood = -317.28198  
Iteration 1:  Log likelihood = -285.39103  
Iteration 2:  Log likelihood = -280.54321  
Iteration 3:  Log likelihood = -279.55563  
Iteration 4:  Log likelihood =  -279.2643  
Iteration 5:  Log likelihood = -279.17041  
Iteration 6:  Log likelihood = -279.13948  
Iteration 7:  Log likelihood = -279.12914  
Iteration 8:  Log likelihood = -279.12564  
Iteration 9:  Log likelihood = -279.12444  
Iteration 10: Log likelihood = -279.12403  
Iteration 11: Log likelihood = -279.12389  
Iteration 12: Log likelihood = -279.12384  
Iteration 13: Log likelihood = -279.12382  
Iteration 14: Log likelihood = -279.12382  
Iteration 15: Log likelihood = -279.12382  
Iteration 16: Log likelihood = -279.12382  
Iteration 17: Log likelihood = -279.12382  
Iteration 18: Log likelihood = -279.12382  
Iteration 19: Log likelihood = -279.12382  (not concave)
Iteration 20: Log likelihood = -279.12382  (not concave)
Iteration 21: Log likelihood = -279.12382  (not concave)
Iteration 22: Log likelihood = -279.12382  (not concave)
Iteration 23: Log likelihood = -279.12382  (not concave)
Iteration 24: Log likelihood = -279.12382  (not concave)
Iteration 25: Log likelihood = -279.12382  (not concave)
Iteration 26: Log likelihood = -279.12382  (not concave)
Iteration 27: Log likelihood = -279.12382  (not concave)
Iteration 28: Log likelihood = -279.12382  (not concave)
Iteration 29: Log likelihood = -279.12382  (not concave)
Iteration 30: Log likelihood = -279.12382  (not concave)
Iteration 31: Log likelihood = -279.12382  (not concave)
Iteration 32: Log likelihood = -279.12382  (not concave)
Iteration 33: Log likelihood = -279.12382  (not concave)
Iteration 34: Log likelihood = -279.12382  (not concave)
Iteration 35: Log likelihood = -279.12382  (not concave)
Iteration 36: Log likelihood = -279.12382  (not concave)
Iteration 37: Log likelihood = -279.12382  (not concave)
Iteration 38: Log likelihood = -279.12382  (not concave)
Iteration 39: Log likelihood = -279.12382  (not concave)
Iteration 40: Log likelihood = -279.12382  (not concave)
Iteration 41: Log likelihood = -279.12382  (not concave)
Iteration 42: Log likelihood = -279.12382  (not concave)
Iteration 43: Log likelihood = -279.12382  (not concave)
Iteration 44: Log likelihood = -279.12382  (not concave)
Iteration 45: Log likelihood = -279.12382  (not concave)
Iteration 46: Log likelihood = -279.12382  (not concave)
Iteration 47: Log likelihood = -279.12382  (not concave)
Iteration 48: Log likelihood = -279.12382  (not concave)
Iteration 49: Log likelihood = -279.12382  (not concave)
Iteration 50: Log likelihood = -279.12382  (not concave)
Iteration 51: Log likelihood = -279.12382  (not concave)
Iteration 52: Log likelihood = -279.12382  (not concave)
Iteration 53: Log likelihood = -279.12382  (not concave)
Iteration 54: Log likelihood = -279.12382  (not concave)
Iteration 55: Log likelihood = -279.12382  (not concave)
Iteration 56: Log likelihood = -279.12382  (not concave)
Iteration 57: Log likelihood = -279.12382  (not concave)
Iteration 58: Log likelihood = -279.12382  (not concave)
Iteration 59: Log likelihood = -279.12382  (not concave)
Iteration 60: Log likelihood = -279.12382  (not concave)
Iteration 61: Log likelihood = -279.12382  (not concave)
Iteration 62: Log likelihood = -279.12382  (not concave)
Iteration 63: Log likelihood = -279.12382  (not concave)
Iteration 64: Log likelihood = -279.12382  (not concave)
Iteration 65: Log likelihood = -279.12382  (not concave)
Iteration 66: Log likelihood = -279.12382  (not concave)
Iteration 67: Log likelihood = -279.12382  (not concave)
Iteration 68: Log likelihood = -279.12382  (not concave)
Iteration 69: Log likelihood = -279.12382  (not concave)
Iteration 70: Log likelihood = -279.12382  (not concave)
Iteration 71: Log likelihood = -279.12382  (not concave)
Iteration 72: Log likelihood = -279.12382  (not concave)
Iteration 73: Log likelihood = -279.12382  (not concave)
Iteration 74: Log likelihood = -279.12382  (not concave)
Iteration 75: Log likelihood = -279.12382  (not concave)
Iteration 76: Log likelihood = -279.12382  (not concave)
Iteration 77: Log likelihood = -279.12382  (not concave)
Iteration 78: Log likelihood = -279.12382  (not concave)
Iteration 79: Log likelihood = -279.12382  (not concave)
Iteration 80: Log likelihood = -279.12382  (not concave)
Iteration 81: Log likelihood = -279.12382  (not concave)
Iteration 82: Log likelihood = -279.12382  (not concave)
Iteration 83: Log likelihood = -279.12382  (not concave)
Iteration 84: Log likelihood = -279.12382  (not concave)
Iteration 85: Log likelihood = -279.12382  (not concave)
Iteration 86: Log likelihood = -279.12382  (not concave)
Iteration 87: Log likelihood = -279.12382  (not concave)
Iteration 88: Log likelihood = -279.12382  (not concave)
Iteration 89: Log likelihood = -279.12382  (not concave)
Iteration 90: Log likelihood = -279.12382  (not concave)
Iteration 91: Log likelihood = -279.12382  (not concave)
Iteration 92: Log likelihood = -279.12382  (not concave)
Iteration 93: Log likelihood = -279.12382  (not concave)
Iteration 94: Log likelihood = -279.12382  (not concave)
Iteration 95: Log likelihood = -279.12382  (not concave)
Iteration 96: Log likelihood = -279.12382  (not concave)
Iteration 97: Log likelihood = -279.12382  (not concave)
Iteration 98: Log likelihood = -279.12382  (not concave)
Iteration 99: Log likelihood = -279.12382  (not concave)
Iteration 100: Log likelihood = -279.12382  (not concave)
Iteration 101: Log likelihood = -279.12382  (not concave)
Iteration 102: Log likelihood = -279.12382  (not concave)
Iteration 103: Log likelihood = -279.12382  (not concave)
Iteration 104: Log likelihood = -279.12382  (not concave)
Iteration 105: Log likelihood = -279.12382  (not concave)
Iteration 106: Log likelihood = -279.12382  (not concave)
Iteration 107: Log likelihood = -279.12382  (not concave)
Iteration 108: Log likelihood = -279.12382  (not concave)
Iteration 109: Log likelihood = -279.12382  (not concave)
Iteration 110: Log likelihood = -279.12382  (not concave)
Iteration 111: Log likelihood = -279.12382  (not concave)
Iteration 112: Log likelihood = -279.12382  (not concave)
Iteration 113: Log likelihood = -279.12382  (not concave)
Iteration 114: Log likelihood = -279.12382  (not concave)
Iteration 115: Log likelihood = -279.12382  (not concave)
Iteration 116: Log likelihood = -279.12382  (not concave)
Iteration 117: Log likelihood = -279.12382  (not concave)
Iteration 118: Log likelihood = -279.12382  (not concave)
Iteration 119: Log likelihood = -279.12382  (not concave)
Iteration 120: Log likelihood = -279.12382  (not concave)
Iteration 121: Log likelihood = -279.12382  (not concave)
Iteration 122: Log likelihood = -279.12382  (not concave)
Iteration 123: Log likelihood = -279.12382  (not concave)
Iteration 124: Log likelihood = -279.12382  (not concave)
Iteration 125: Log likelihood = -279.12382  (not concave)
Iteration 126: Log likelihood = -279.12382  (not concave)
Iteration 127: Log likelihood = -279.12382  (not concave)
Iteration 128: Log likelihood = -279.12382  (not concave)
Iteration 129: Log likelihood = -279.12382  (not concave)
Iteration 130: Log likelihood = -279.12382  (not concave)
Iteration 131: Log likelihood = -279.12382  (not concave)
Iteration 132: Log likelihood = -279.12382  (not concave)
Iteration 133: Log likelihood = -279.12382  (not concave)
Iteration 134: Log likelihood = -279.12382  (not concave)
Iteration 135: Log likelihood = -279.12382  (not concave)
Iteration 136: Log likelihood = -279.12382  (not concave)
Iteration 137: Log likelihood = -279.12382  (not concave)
Iteration 138: Log likelihood = -279.12382  (not concave)
Iteration 139: Log likelihood = -279.12382  (not concave)
Iteration 140: Log likelihood = -279.12382  (not concave)
Iteration 141: Log likelihood = -279.12382  (not concave)
Iteration 142: Log likelihood = -279.12382  (not concave)
Iteration 143: Log likelihood = -279.12382  (not concave)
Iteration 144: Log likelihood = -279.12382  (not concave)
Iteration 145: Log likelihood = -279.12382  (not concave)
Iteration 146: Log likelihood = -279.12382  (not concave)
Iteration 147: Log likelihood = -279.12382  (not concave)
Iteration 148: Log likelihood = -279.12382  (not concave)
Iteration 149: Log likelihood = -279.12382  (not concave)
Iteration 150: Log likelihood = -279.12382  (not concave)
Iteration 151: Log likelihood = -279.12382  (not concave)
Iteration 152: Log likelihood = -279.12382  (not concave)
Iteration 153: Log likelihood = -279.12382  (not concave)
Iteration 154: Log likelihood = -279.12382  (not concave)
Iteration 155: Log likelihood = -279.12382  (not concave)
Iteration 156: Log likelihood = -279.12382  (not concave)
Iteration 157: Log likelihood = -279.12382  (not concave)
Iteration 158: Log likelihood = -279.12382  (not concave)
Iteration 159: Log likelihood = -279.12382  (not concave)
Iteration 160: Log likelihood = -279.12382  (not concave)
Iteration 161: Log likelihood = -279.12382  (not concave)
Iteration 162: Log likelihood = -279.12382  (not concave)
Iteration 163: Log likelihood = -279.12382  (not concave)
Iteration 164: Log likelihood = -279.12382  (not concave)
Iteration 165: Log likelihood = -279.12382  (not concave)
Iteration 166: Log likelihood = -279.12382  (not concave)
Iteration 167: Log likelihood = -279.12382  (not concave)
Iteration 168: Log likelihood = -279.12382  (not concave)
Iteration 169: Log likelihood = -279.12382  (not concave)
Iteration 170: Log likelihood = -279.12382  (not concave)
Iteration 171: Log likelihood = -279.12382  (not concave)
Iteration 172: Log likelihood = -279.12382  (not concave)
Iteration 173: Log likelihood = -279.12382  (not concave)
Iteration 174: Log likelihood = -279.12382  (not concave)
Iteration 175: Log likelihood = -279.12382  (not concave)
Iteration 176: Log likelihood = -279.12382  (not concave)
Iteration 177: Log likelihood = -279.12382  (not concave)
Iteration 178: Log likelihood = -279.12382  (not concave)
Iteration 179: Log likelihood = -279.12382  (not concave)
Iteration 180: Log likelihood = -279.12382  (not concave)
Iteration 181: Log likelihood = -279.12382  (not concave)
Iteration 182: Log likelihood = -279.12382  (not concave)
Iteration 183: Log likelihood = -279.12382  (not concave)
Iteration 184: Log likelihood = -279.12382  (not concave)
Iteration 185: Log likelihood = -279.12382  (not concave)
Iteration 186: Log likelihood = -279.12382  (not concave)
Iteration 187: Log likelihood = -279.12382  (not concave)
Iteration 188: Log likelihood = -279.12382  (not concave)
Iteration 189: Log likelihood = -279.12382  (not concave)
Iteration 190: Log likelihood = -279.12382  (not concave)
Iteration 191: Log likelihood = -279.12382  (not concave)
Iteration 192: Log likelihood = -279.12382  (not concave)
Iteration 193: Log likelihood = -279.12382  (not concave)
Iteration 194: Log likelihood = -279.12382  (not concave)
Iteration 195: Log likelihood = -279.12382  (not concave)
Iteration 196: Log likelihood = -279.12382  (not concave)
Iteration 197: Log likelihood = -279.12382  (not concave)
Iteration 198: Log likelihood = -279.12382  (not concave)
Iteration 199: Log likelihood = -279.12382  (not concave)
Iteration 200: Log likelihood = -279.12382  (not concave)
Iteration 201: Log likelihood = -279.12382  (not concave)
Iteration 202: Log likelihood = -279.12382  (not concave)
Iteration 203: Log likelihood = -279.12382  (not concave)
Iteration 204: Log likelihood = -279.12382  (not concave)
Iteration 205: Log likelihood = -279.12382  (not concave)
Iteration 206: Log likelihood = -279.12382  (not concave)
Iteration 207: Log likelihood = -279.12382  (not concave)
Iteration 208: Log likelihood = -279.12382  (not concave)
Iteration 209: Log likelihood = -279.12382  (not concave)
Iteration 210: Log likelihood = -279.12382  (not concave)
Iteration 211: Log likelihood = -279.12382  (not concave)
Iteration 212: Log likelihood = -279.12382  (not concave)
Iteration 213: Log likelihood = -279.12382  (not concave)
Iteration 214: Log likelihood = -279.12382  (not concave)
Iteration 215: Log likelihood = -279.12382  (not concave)
Iteration 216: Log likelihood = -279.12382  (not concave)
Iteration 217: Log likelihood = -279.12382  (not concave)
Iteration 218: Log likelihood = -279.12382  (not concave)
Iteration 219: Log likelihood = -279.12382  (not concave)
Iteration 220: Log likelihood = -279.12382  (not concave)
Iteration 221: Log likelihood = -279.12382  (not concave)
Iteration 222: Log likelihood = -279.12382  (not concave)
Iteration 223: Log likelihood = -279.12382  (not concave)
Iteration 224: Log likelihood = -279.12382  (not concave)
Iteration 225: Log likelihood = -279.12382  (not concave)
Iteration 226: Log likelihood = -279.12382  (not concave)
Iteration 227: Log likelihood = -279.12382  (not concave)
Iteration 228: Log likelihood = -279.12382  (not concave)
Iteration 229: Log likelihood = -279.12382  (not concave)
Iteration 230: Log likelihood = -279.12382  (not concave)
Iteration 231: Log likelihood = -279.12382  (not concave)
Iteration 232: Log likelihood = -279.12382  (not concave)
Iteration 233: Log likelihood = -279.12382  (not concave)
Iteration 234: Log likelihood = -279.12382  (not concave)
Iteration 235: Log likelihood = -279.12382  (not concave)
Iteration 236: Log likelihood = -279.12382  (not concave)
Iteration 237: Log likelihood = -279.12382  (not concave)
Iteration 238: Log likelihood = -279.12382  (not concave)
Iteration 239: Log likelihood = -279.12382  (not concave)
Iteration 240: Log likelihood = -279.12382  (not concave)
Iteration 241: Log likelihood = -279.12382  (not concave)
Iteration 242: Log likelihood = -279.12382  (not concave)
Iteration 243: Log likelihood = -279.12382  (not concave)
Iteration 244: Log likelihood = -279.12382  (not concave)
Iteration 245: Log likelihood = -279.12382  (not concave)
Iteration 246: Log likelihood = -279.12382  (not concave)
Iteration 247: Log likelihood = -279.12382  (not concave)
Iteration 248: Log likelihood = -279.12382  (not concave)
Iteration 249: Log likelihood = -279.12382  (not concave)
Iteration 250: Log likelihood = -279.12382  (not concave)
Iteration 251: Log likelihood = -279.12382  (not concave)
Iteration 252: Log likelihood = -279.12382  (not concave)
Iteration 253: Log likelihood = -279.12382  (not concave)
Iteration 254: Log likelihood = -279.12382  (not concave)
Iteration 255: Log likelihood = -279.12382  (not concave)
Iteration 256: Log likelihood = -279.12382  (not concave)
Iteration 257: Log likelihood = -279.12382  (not concave)
Iteration 258: Log likelihood = -279.12382  (not concave)
Iteration 259: Log likelihood = -279.12382  (not concave)
Iteration 260: Log likelihood = -279.12382  (not concave)
Iteration 261: Log likelihood = -279.12382  (not concave)
Iteration 262: Log likelihood = -279.12382  (not concave)
Iteration 263: Log likelihood = -279.12382  (not concave)
Iteration 264: Log likelihood = -279.12382  (not concave)
Iteration 265: Log likelihood = -279.12382  (not concave)
Iteration 266: Log likelihood = -279.12382  (not concave)
Iteration 267: Log likelihood = -279.12382  (not concave)
Iteration 268: Log likelihood = -279.12382  (not concave)
Iteration 269: Log likelihood = -279.12382  (not concave)
Iteration 270: Log likelihood = -279.12382  (not concave)
Iteration 271: Log likelihood = -279.12382  (not concave)
Iteration 272: Log likelihood = -279.12382  (not concave)
Iteration 273: Log likelihood = -279.12382  (not concave)
Iteration 274: Log likelihood = -279.12382  (not concave)
Iteration 275: Log likelihood = -279.12382  (not concave)
Iteration 276: Log likelihood = -279.12382  (not concave)
Iteration 277: Log likelihood = -279.12382  (not concave)
Iteration 278: Log likelihood = -279.12382  (not concave)
Iteration 279: Log likelihood = -279.12382  (not concave)
Iteration 280: Log likelihood = -279.12382  (not concave)
Iteration 281: Log likelihood = -279.12382  (not concave)
Iteration 282: Log likelihood = -279.12382  (not concave)
Iteration 283: Log likelihood = -279.12382  (not concave)
Iteration 284: Log likelihood = -279.12382  (not concave)
Iteration 285: Log likelihood = -279.12382  (not concave)
Iteration 286: Log likelihood = -279.12382  (not concave)
Iteration 287: Log likelihood = -279.12382  (not concave)
Iteration 288: Log likelihood = -279.12382  (not concave)
Iteration 289: Log likelihood = -279.12382  (not concave)
Iteration 290: Log likelihood = -279.12382  (not concave)
Iteration 291: Log likelihood = -279.12382  (not concave)
Iteration 292: Log likelihood = -279.12382  (not concave)
Iteration 293: Log likelihood = -279.12382  (not concave)
Iteration 294: Log likelihood = -279.12382  (not concave)
Iteration 295: Log likelihood = -279.12382  (not concave)
Iteration 296: Log likelihood = -279.12382  (not concave)
Iteration 297: Log likelihood = -279.12382  (not concave)
Iteration 298: Log likelihood = -279.12382  (not concave)
Iteration 299: Log likelihood = -279.12382  (not concave)
Iteration 300: Log likelihood = -279.12382  (not concave)
convergence not achieved

Probit regression                                       Number of obs =  3,838
                                                        LR chi2(37)   =  76.32
                                                        Prob > chi2   = 0.0002
Log likelihood = -279.12382                             Pseudo R2     = 0.1203

------------------------------------------------------------------------------------------
    hard_final_Exact_new | Coefficient  Std. err.      z    P>|z|     [95% conf. interval]
-------------------------+----------------------------------------------------------------
             Firm_Size_w |  -.1331548   .0879856    -1.51   0.130    -.3056033    .0392938
                   ROA_w |   1.753219   1.653727     1.06   0.289    -1.488025    4.994464
              Leverage_w |  -.8882499   .5285162    -1.68   0.093    -1.924123    .1476229
      Market_book_four_w |   .0055092   .0083093     0.66   0.507    -.0107767    .0217951
       Non_pension_CFO_w |  -3.671549   1.940931    -1.89   0.059    -7.475703    .1326055
                 STD_CFO |   4.207263   3.133094     1.34   0.179    -1.933487    10.34801
            board_size_w |   .0175808   .0305472     0.58   0.565    -.0422906    .0774522
GenderRatiogenderratio_w |  -.1347727   .7153949    -0.19   0.851    -1.536921    1.267376
independent_percentage_w |   .5566361   .7793484     0.71   0.475    -.9708587    2.084131
           Fund_Status_w |  -1.613321    2.29322    -0.70   0.482    -6.107949    2.881307
         FUNDING_RATIO_w |  -.6834395   .5025799    -1.36   0.174    -1.668478    .3015989
            Platn_Size_w |   .0959977   .0768274     1.25   0.211    -.0545812    .2465766
                         |
                    year |
                   2005  |   -91.8798          .        .       .            .           .
                   2006  |  -93.79258          .        .       .            .           .
                   2007  |   4.281815   1.262739     3.39   0.001     1.806892    6.756738
                   2008  |   4.691972   1.229545     3.82   0.000     2.282107    7.101836
                   2009  |   5.098007   1.225024     4.16   0.000     2.697005    7.499009
                   2010  |    4.96413   1.227161     4.05   0.000     2.558939    7.369321
                   2011  |   4.706775   1.240509     3.79   0.000     2.275421    7.138129
                   2012  |   4.929779   1.227125     4.02   0.000     2.524658    7.334899
                   2013  |   4.691142    1.23845     3.79   0.000     2.263826    7.118459
                   2014  |   4.911324   1.230808     3.99   0.000     2.498985    7.323663
                   2015  |   5.026041   1.230451     4.08   0.000     2.614402     7.43768
                   2016  |   4.756492   1.237431     3.84   0.000     2.331173    7.181812
                   2017  |   4.962073   1.218948     4.07   0.000     2.572979    7.351167
                   2018  |   5.291178   1.215679     4.35   0.000      2.90849    7.673866
                   2019  |   4.786223   1.228021     3.90   0.000     2.379346      7.1931
                   2020  |   5.314354   1.199832     4.43   0.000     2.962727    7.665982
                   2021  |   4.892568   1.222439     4.00   0.000     2.496632    7.288504
                   2022  |   4.624751   1.248862     3.70   0.000     2.177026    7.072476
                         |
                   ff_12 |
                      2  |  -.2459353   .4661325    -0.53   0.598    -1.159538    .6676676
                      3  |  -.0980302   .2452122    -0.40   0.689    -.5786372    .3825769
                      4  |  -.3404774   .4167352    -0.82   0.414    -1.157263    .4763087
                      5  |  -.1001264   .2845053    -0.35   0.725    -.6577465    .4574937
                      6  |  -.0363145   .2589761    -0.14   0.888    -.5438983    .4712693
                      7  |   -4.55164          .        .       .            .           .
                      8  |  -.3337711    .290807    -1.15   0.251    -.9037424    .2362002
                      9  |   .5119286     .29496     1.74   0.083    -.0661823    1.090039
                     10  |  -.0832234   .2705026    -0.31   0.758    -.6133988     .446952
                     11  |   .2711055    .276386     0.98   0.327    -.2706011    .8128121
                     12  |  -.2878671   .2936605    -0.98   0.327    -.8634311    .2876969
                         |
                   _cons |  -6.107303          .        .       .            .           .
------------------------------------------------------------------------------------------
Note: 693 failures and 0 successes completely determined.
Warning: Convergence not achieved.
note: two-step estimate of rho = 1.0681901 is being truncated to 1
year: factor variable base category conflict
r(198);

      Comment

      Previous Next
      Working...
      X