mmqreg when X!=Z - Statalist

Fadi Ansar

Join Date: May 2021
Posts: 119

#61

13 Sep 2021, 10:02

Dear Joao Santos Silva ,here is the results with xtreg .I am having observations for that years as well ,Can I consider a model like this The data for CBT variable available till 2014

Code:

 xtreg Zscore d_MP  CBT  NIM   lasset CapitalRatio ownership#C.d_MP i.Year ,fe robust
note: 2014.Year omitted because of collinearity

Fixed-effects (within) regression               Number of obs     =        645
Group variable: bankname                        Number of groups  =         85

R-sq:                                           Obs per group:
     within  = 0.9401                                         min =          1
     between = 0.9702                                         avg =        7.6
     overall = 0.9393                                         max =          9

                                                F(14,84)          =   14940.37
corr(u_i, Xb)  = -0.0436                        Prob > F          =     0.0000

                                  (Std. Err. adjusted for 85 clusters in bankname)
----------------------------------------------------------------------------------
                 |               Robust
          Zscore |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-----------------+----------------------------------------------------------------
            d_MP |    .029925   .0176345     1.70   0.093    -.0051432    .0649932
             CBT |   .6605285   .0150756    43.81   0.000     .6305491     .690508
             NIM |   .0279845   .0186889     1.50   0.138    -.0091804    .0651494
          lasset |   .0006398   .0038937     0.16   0.870    -.0071032    .0083827
    CapitalRatio |  -.0040615   .0024499    -1.66   0.101    -.0089333    .0008104
                 |
ownership#c.d_MP |
 Private Sector  |  -.0183861   .0117396    -1.57   0.121    -.0417316    .0049593
  Public Sector  |  -.0236916   .0150531    -1.57   0.119    -.0536265    .0062432
                 |
            Year |
           2007  |  -.4461567   .0707734    -6.30   0.000    -.5868974   -.3054159
           2008  |    1.87928   .0372979    50.39   0.000     1.805109    1.953451
           2009  |   1.249165   .0458205    27.26   0.000     1.158046    1.340284
           2010  |   1.695459   .0179887    94.25   0.000     1.659687    1.731232
           2011  |   1.417045   .0199934    70.88   0.000     1.377286    1.456804
           2012  |   1.564757   .0226365    69.13   0.000     1.519742    1.609772
           2013  |   1.688618   .0159957   105.57   0.000     1.656809    1.720428
           2014  |          0  (omitted)
                 |
           _cons |    13.4738   .1052384   128.03   0.000     13.26452    13.68308
-----------------+----------------------------------------------------------------
         sigma_u |  .04461904
         sigma_e |  .20766117
             rho |  .04412949   (fraction of variance due to u_i)
----------------------------------------------------------------------------------

Last edited by Fadi Ansar; 13 Sep 2021, 10:09.

Comment

Fadi Ansar

Join Date: May 2021
Posts: 119

#62

13 Sep 2021, 10:07

I am adding one more thing I noticed without CBT variable the result is like this

Code:

 xtreg Zscore d_MP    NIM   lasset CapitalRatio ownership#C.d_MP i.Year ,fe robust

Fixed-effects (within) regression               Number of obs     =      1,009
Group variable: bankname                        Number of groups  =         96

R-sq:                                           Obs per group:
     within  = 0.9543                                         min =          1
     between = 0.9747                                         avg =       10.5
     overall = 0.9508                                         max =         14

                                                F(19,95)          =   40934.94
corr(u_i, Xb)  = -0.0392                        Prob > F          =     0.0000

                                  (Std. Err. adjusted for 96 clusters in bankname)
----------------------------------------------------------------------------------
                 |               Robust
          Zscore |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-----------------+----------------------------------------------------------------
            d_MP |   .0243656   .0145675     1.67   0.098    -.0045546    .0532858
             NIM |    .005412   .0040629     1.33   0.186     -.002654    .0134779
          lasset |   .0187463   .0132377     1.42   0.160    -.0075338    .0450264
    CapitalRatio |   .0021085    .001767     1.19   0.236    -.0013995    .0056164
                 |
ownership#c.d_MP |
 Private Sector  |  -.0154107   .0101904    -1.51   0.134    -.0356413    .0048199
  Public Sector  |  -.0200839   .0128017    -1.57   0.120    -.0454985    .0053307
                 |
            Year |
           2007  |  -.4550871   .0652665    -6.97   0.000    -.5846576   -.3255166
           2008  |   1.872348   .0378236    49.50   0.000     1.797258    1.947437
           2009  |   1.225587   .0314412    38.98   0.000     1.163168    1.288005
           2010  |   1.682852   .0229587    73.30   0.000     1.637273    1.728431
           2011  |   1.405687   .0158297    88.80   0.000     1.374261    1.437113
           2012  |    1.54855   .0147194   105.20   0.000     1.519328    1.577772
           2013  |   1.667222   .0230442    72.35   0.000     1.621473     1.71297
           2014  |   .9622465   .0147375    65.29   0.000     .9329889    .9915042
           2015  |   .9165596   .0240532    38.11   0.000     .8688081    .9643112
           2016  |   1.940814   .0387458    50.09   0.000     1.863894    2.017734
           2017  |   .4040081    .013806    29.26   0.000     .3765996    .4314165
           2018  |   1.278672   .0400392    31.94   0.000     1.199185     1.35816
           2019  |   3.888779   .0129997   299.14   0.000     3.862972    3.914587
                 |
           _cons |   15.35831    .146116   105.11   0.000     15.06823    15.64839
-----------------+----------------------------------------------------------------
         sigma_u |  .07161595
         sigma_e |  .22168762
             rho |  .09449867   (fraction of variance due to u_i)
----------------------------------------------------------------------------------

.

for xtqreg with out CBT variable

Code:

 xtqreg Zscore d_MP    NIM   lasset CapitalRatio ownership#C.d_MP i.Year ,quantile(.1 .2 .3 .4 .5
> .6 .7 .8 .9 )



                              MM-QR regression results
Number of obs = 1009
WARNING: some fitted values of the scale function are negative
.1 Quantile regression
----------------------------------------------------------------------------------
                 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-----------------+----------------------------------------------------------------
            d_MP |   .0337665   1.337457     0.03   0.980    -2.587602    2.655135
             NIM |  -.0241425   .9659364    -0.02   0.980    -1.917343    1.869058
          lasset |  -.0048588   .7077136    -0.01   0.995    -1.391952    1.382234
    CapitalRatio |  -.0003297   .1652998    -0.00   0.998    -.3243114     .323652
                 |
ownership#c.d_MP |
 Private Sector  |  -.0224556   1.044202    -0.02   0.983    -2.069054    2.024143
  Public Sector  |  -.0414594   1.223653    -0.03   0.973    -2.439776    2.356857
                 |
            Year |
           2005  |          0  (empty)
           2006  |  -3.894701   3.367319    -1.16   0.247    -10.49453    2.705124
           2007  |  -4.597462   4.499426    -1.02   0.307    -13.41618    4.221251
           2008  |  -2.094676    3.35129    -0.63   0.532    -8.663083    4.473732
           2009  |  -2.609309   4.764655    -0.55   0.584    -11.94786    6.729243
           2010  |  -2.171245   .1645113   -13.20   0.000    -2.493681   -1.848809
           2011  |  -2.404319   3.218529    -0.75   0.455     -8.71252    3.903881
           2012  |   -2.23801   3.375451    -0.66   0.507    -8.853772    4.377752
           2013  |  -2.205338   3.263159    -0.68   0.499    -8.601011    4.190336
           2014  |  -2.841589   3.300299    -0.86   0.389    -9.310057    3.626879
           2015  |  -2.925681   3.422416    -0.85   0.393    -9.633492     3.78213
           2016  |  -2.028348   3.921632    -0.52   0.605    -9.714606     5.65791
           2017  |  -3.430556   3.331702    -1.03   0.303    -9.960572     3.09946
           2018  |  -2.755448   3.393795    -0.81   0.417    -9.407164    3.896269
           2019  |          0   2.634764     0.00   1.000    -5.164042    5.164042
----------------------------------------------------------------------------------

.2 Quantile regression
----------------------------------------------------------------------------------
                 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-----------------+----------------------------------------------------------------
            d_MP |   .0295655    .927837     0.03   0.975    -1.788962    1.848093
             NIM |  -.0109355   .6701117    -0.02   0.987     -1.32433    1.302459
          lasset |   .0056896    .490972     0.01   0.991    -.9565977     .967977
    CapitalRatio |   .0007598   .1146743     0.01   0.995    -.2239976    .2255173
                 |
ownership#c.d_MP |
 Private Sector  |  -.0193074    .724396    -0.03   0.979    -1.439097    1.400483
  Public Sector  |  -.0319073   .8488919    -0.04   0.970    -1.695705     1.63189
                 |
            Year |
           2005  |          0  (empty)
           2006  |  -3.892055   2.336015    -1.67   0.096    -8.470559    .6864497
           2007  |  -4.484137   3.121508    -1.44   0.151    -10.60218    1.633906
           2008  |  -2.059711   2.324911    -0.89   0.376    -6.616452    2.497031
           2009  |  -2.633388   3.305413    -0.80   0.426    -9.111879    3.845102
           2010  |  -2.186743   .1143954   -19.12   0.000    -2.410954   -1.962533
           2011  |  -2.439521   2.232853    -1.09   0.275    -6.815832     1.93679
           2012  |  -2.283689   2.341751    -0.98   0.329    -6.873436    2.306058
           2013  |  -2.212586   2.263761    -0.98   0.328    -6.649476    2.224304
           2014  |  -2.879548    2.28959    -1.26   0.209    -7.367062    1.607965
           2015  |  -2.946478   2.374259    -1.24   0.215     -7.59994    1.706985
           2016  |  -1.992427   2.720581    -0.73   0.464    -7.324669    3.339814
           2017  |  -3.454783    2.31134    -1.49   0.135    -7.984927    1.075361
           2018  |  -2.690499   2.354448    -1.14   0.253    -7.305132    1.924134
           2019  |          0    1.82782     0.00   1.000     -3.58246     3.58246
----------------------------------------------------------------------------------

.3 Quantile regression
----------------------------------------------------------------------------------
                 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-----------------+----------------------------------------------------------------
            d_MP |   .0275473   .7391159     0.04   0.970    -1.421093    1.476188
             NIM |  -.0045905   .5338168    -0.01   0.993    -1.050852    1.041671
          lasset |   .0107573   .3911126     0.03   0.978    -.7558093    .7773239
    CapitalRatio |   .0012833   .0913499     0.01   0.989    -.1777592    .1803258
                 |
ownership#c.d_MP |
 Private Sector  |   -.017795   .5770544    -0.03   0.975    -1.148801    1.113211
  Public Sector  |  -.0273183   .6762303    -0.04   0.968    -1.352705    1.298069
                 |
            Year |
           2005  |          0  (empty)
           2006  |  -3.890783    1.86087    -2.09   0.037    -7.538022   -.2435447
           2007  |  -4.429694   2.486642    -1.78   0.075    -9.303422    .4440339
           2008  |  -2.042913   1.852032    -1.10   0.270    -5.672829    1.587004
           2009  |  -2.644956   2.633106    -1.00   0.315     -7.80575    2.515838
           2010  |  -2.194189   .0912782   -24.04   0.000    -2.373091   -2.015287
           2011  |  -2.456432   1.778724    -1.38   0.167    -5.942668    1.029803
           2012  |  -2.305634   1.865492    -1.24   0.216    -5.961931    1.350663
           2013  |  -2.216068   1.803316    -1.23   0.219    -5.750503    1.318367
           2014  |  -2.897784   1.823927    -1.59   0.112    -6.472615    .6770463
           2015  |  -2.956469   1.891349    -1.56   0.118    -6.663445    .7505066
           2016  |   -1.97517   2.167228    -0.91   0.362    -6.222858    2.272518
           2017  |  -3.466422   1.841234    -1.88   0.060    -7.075175    .1423308
           2018  |  -2.659296   1.875584    -1.42   0.156    -6.335373    1.016781
           2019  |          0   1.456043     0.00   1.000    -2.853791    2.853791
----------------------------------------------------------------------------------

.4 Quantile regression
----------------------------------------------------------------------------------
                 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-----------------+----------------------------------------------------------------
            d_MP |   .0266596   .6594653     0.04   0.968    -1.265869    1.319188
             NIM |     -.0018   .4762892    -0.00   0.997    -.9353098    .9317098
          lasset |   .0129861   .3489639     0.04   0.970    -.6709705    .6969427
    CapitalRatio |   .0015135   .0815055     0.02   0.985    -.1582344    .1612614
                 |
ownership#c.d_MP |
 Private Sector  |  -.0171298   .5148683    -0.03   0.973    -1.026253    .9919936
  Public Sector  |     -.0253   .6033561    -0.04   0.967    -1.207856    1.157256
                 |
            Year |
           2005  |          0  (empty)
           2006  |  -3.890224   1.660334    -2.34   0.019     -7.14442   -.6360286
           2007  |  -4.405749   2.218666    -1.99   0.047    -8.754255   -.0572437
           2008  |  -2.035525   1.652448    -1.23   0.218    -5.274263    1.203214
           2009  |  -2.650044   2.349347    -1.13   0.259     -7.25468    1.954593
           2010  |  -2.197464   .0814014   -27.00   0.000    -2.357008    -2.03792
           2011  |   -2.46387   1.587032    -1.55   0.121    -5.574396    .6466563
           2012  |  -2.315285   1.664445    -1.39   0.164    -5.577537     .946966
           2013  |    -2.2176   1.608982    -1.38   0.168    -5.371146    .9359467
           2014  |  -2.905805   1.627362    -1.79   0.074    -6.095376    .2837663
           2015  |  -2.960863   1.687525    -1.75   0.079    -6.268352    .3466253
           2016  |  -1.967581   1.933675    -1.02   0.309    -5.757514    1.822353
           2017  |  -3.471541   1.642809    -2.11   0.035    -6.691389   -.2516942
           2018  |  -2.645573   1.673458    -1.58   0.114    -5.925491    .6343442
           2019  |          0   1.299133     0.00   1.000    -2.546253    2.546253
----------------------------------------------------------------------------------

.5 Quantile regression
----------------------------------------------------------------------------------
                 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-----------------+----------------------------------------------------------------
            d_MP |   .0245097   .4830541     0.05   0.960    -.9222589    .9712784
             NIM |   .0049588   .3488777     0.01   0.989     -.678829    .6887466
          lasset |   .0183843   .2556135     0.07   0.943    -.4826089    .5193776
    CapitalRatio |   .0020711   .0597022     0.03   0.972     -.114943    .1190852
                 |
ownership#c.d_MP |
 Private Sector  |  -.0155187    .377138    -0.04   0.967    -.7546956    .7236582
  Public Sector  |  -.0204116   .4419538    -0.05   0.963    -.8866251    .8458019
                 |
            Year |
           2005  |          0  (empty)
           2006  |   -3.88887   1.216185    -3.20   0.001    -6.272549   -1.505191
           2007  |  -4.347755   1.625181    -2.68   0.007    -7.533051   -1.162458
           2008  |  -2.017631    1.21041    -1.67   0.096    -4.389991    .3547285
           2009  |  -2.662367   1.720872    -1.55   0.122    -6.035214    .7104811
           2010  |  -2.205396   .0595027   -37.06   0.000    -2.322019   -2.088773
           2011  |  -2.481885   1.162467    -2.14   0.033    -4.760278   -.2034915
           2012  |  -2.338662   1.219156    -1.92   0.055    -4.728163    .0508389
           2013  |  -2.221309   1.178566    -1.88   0.059    -4.531257    .0886386
           2014  |   -2.92523   1.192003    -2.45   0.014    -5.261514    -.588947
           2015  |  -2.971506   1.236092    -2.40   0.016    -5.394201    -.548811
           2016  |  -1.949198   1.416402    -1.38   0.169    -4.725294    .8268986
           2017  |   -3.48394   1.203332    -2.90   0.004    -5.842427   -1.125453
           2018  |  -2.612335   1.225797    -2.13   0.033    -5.014854   -.2098167
           2019  |          0    .951606     0.00   1.000    -1.865113    1.865113
----------------------------------------------------------------------------------

.6 Quantile regression
----------------------------------------------------------------------------------
                 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-----------------+----------------------------------------------------------------
            d_MP |   .0220189   .3473416     0.06   0.949    -.6587581    .7027958
             NIM |   .0127895   .2508564     0.05   0.959    -.4788799     .504459
          lasset |   .0246387   .1837956     0.13   0.893    -.3355941    .3848715
    CapitalRatio |   .0027171   .0429288     0.06   0.950    -.0814218     .086856
                 |
ownership#c.d_MP |
 Private Sector  |  -.0136521   .2711825    -0.05   0.960    -.5451601    .5178558
  Public Sector  |   -.014748    .317786    -0.05   0.963    -.6375971    .6081012
                 |
            Year |
           2005  |          0  (empty)
           2006  |  -3.887301   .8745031    -4.45   0.000    -5.601296   -2.173306
           2007  |  -4.280562    1.16854    -3.66   0.000    -6.570859   -1.990266
           2008  |    -1.9969   .8703427    -2.29   0.022     -3.70274   -.2910596
           2009  |  -2.676643   1.237388    -2.16   0.031     -5.10188   -.2514069
           2010  |  -2.214585   .0426422   -51.93   0.000    -2.298162   -2.131008
           2011  |  -2.502756   .8358456    -2.99   0.003    -4.140983   -.8645288
           2012  |  -2.365746   .8765888    -2.70   0.007    -4.083829   -.6476635
           2013  |  -2.225607   .8474501    -2.63   0.009    -3.886578    -.564635
           2014  |  -2.947737   .8570784    -3.44   0.001     -4.62758   -1.267894
           2015  |  -2.983837   .8888049    -3.36   0.001    -4.725862   -1.241811
           2016  |    -1.9279   1.018458    -1.89   0.058    -3.924042    .0682419
           2017  |  -3.498305   .8652422    -4.04   0.000    -5.194148   -1.802461
           2018  |  -2.573826   .8813829    -2.92   0.003    -4.301305   -.8463472
           2019  |          0   .6842555     0.00   1.000    -1.341116    1.341116
----------------------------------------------------------------------------------

.7 Quantile regression
----------------------------------------------------------------------------------
                 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-----------------+----------------------------------------------------------------
            d_MP |   .0203408   .3432864     0.06   0.953    -.6524881    .6931698
             NIM |   .0180649   .2479242     0.07   0.942    -.4678576    .5039874
          lasset |   .0288521   .1816467     0.16   0.874    -.3271688    .3848731
    CapitalRatio |   .0031523   .0424275     0.07   0.941    -.0800041    .0863087
                 |
ownership#c.d_MP |
 Private Sector  |  -.0123946   .2680164    -0.05   0.963    -.5376972    .5129079
  Public Sector  |  -.0109325   .3140747    -0.03   0.972    -.6265077    .6046426
                 |
            Year |
           2005  |          0  (empty)
           2006  |  -3.886244   .8642939    -4.50   0.000    -5.580229   -2.192259
           2007  |  -4.235296   1.154833    -3.67   0.000    -6.498727   -1.971865
           2008  |  -1.982934   .8601739    -2.31   0.021    -3.668844   -.2970238
           2009  |  -2.686261   1.222942    -2.20   0.028    -5.083184   -.2893385
           2010  |  -2.220776   .0421381   -52.70   0.000    -2.303365   -2.138187
           2011  |  -2.516817   .8260845    -3.05   0.002    -4.135913   -.8977208
           2012  |  -2.383992   .8663502    -2.75   0.006    -4.082007   -.6859766
           2013  |  -2.228502    .837557    -2.66   0.008    -3.870084   -.5869203
           2014  |  -2.962899   .8470693    -3.50   0.000    -4.623124   -1.302674
           2015  |  -2.992144   .8784281    -3.41   0.001    -4.713831   -1.270456
           2016  |  -1.913552   1.006562    -1.90   0.057    -3.886378    .0592741
           2017  |  -3.507982   .8551407    -4.10   0.000    -5.184027   -1.831937
           2018  |  -2.547883    .871067    -2.93   0.003    -4.255143   -.8406233
           2019  |          0   .6762675     0.00   1.000     -1.32546     1.32546
----------------------------------------------------------------------------------

.8 Quantile regression
----------------------------------------------------------------------------------
                 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-----------------+----------------------------------------------------------------
            d_MP |   .0182721   .4420501     0.04   0.967    -.8481301    .8846744
             NIM |   .0245685   .3192481     0.08   0.939    -.6011463    .6502833
          lasset |   .0340465   .2339036     0.15   0.884    -.4243962    .4924892
    CapitalRatio |   .0036888   .0546337     0.07   0.946    -.1033913    .1107689
                 |
ownership#c.d_MP |
 Private Sector  |  -.0108444   .3451252    -0.03   0.975    -.6872773    .6655885
  Public Sector  |  -.0062288   .4044324    -0.02   0.988    -.7989016    .7864441
                 |
            Year |
           2005  |          0  (empty)
           2006  |  -3.884941   1.112953    -3.49   0.000    -6.066289   -1.703593
           2007  |  -4.179491   1.487052    -2.81   0.005     -7.09406   -1.264923
           2008  |  -1.965716   1.107643    -1.77   0.076    -4.136656    .2052245
           2009  |  -2.698118   1.574773    -1.71   0.087    -5.784616    .3883792
           2010  |  -2.228408   .0541152   -41.18   0.000    -2.334472   -2.122344
           2011  |  -2.534151    1.06372    -2.38   0.017    -4.619004   -.4492985
           2012  |  -2.406486   1.115551    -2.16   0.031    -4.592925   -.2200456
           2013  |  -2.232071   1.078521    -2.07   0.038    -4.345933   -.1182098
           2014  |  -2.981591   1.090736    -2.73   0.006    -5.119395    -.843787
           2015  |  -3.002385   1.131141    -2.65   0.008    -5.219381   -.7853886
           2016  |  -1.895863   1.296143    -1.46   0.144    -4.436257    .6445304
           2017  |  -3.519912   1.101147    -3.20   0.001    -5.678122   -1.361703
           2018  |    -2.5159   1.121652    -2.24   0.025    -4.714299    -.317502
           2019  |          0   .8708301     0.00   1.000    -1.706796    1.706796
----------------------------------------------------------------------------------

.9 Quantile regression
----------------------------------------------------------------------------------
                 |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-----------------+----------------------------------------------------------------
            d_MP |   .0130621   .8872025     0.01   0.988    -1.725823    1.751947
             NIM |   .0409477   .6407386     0.06   0.949    -1.214877    1.296772
          lasset |   .0471286   .4694516     0.10   0.920    -.8729796    .9672367
    CapitalRatio |   .0050401   .1096508     0.05   0.963    -.2098715    .2199516
                 |
ownership#c.d_MP |
 Private Sector  |  -.0069401   .6926729    -0.01   0.992    -1.364554    1.350674
  Public Sector  |   .0056177   .8117032     0.01   0.994    -1.585291    1.596527
                 |
            Year |
           2005  |          0  (empty)
           2006  |  -3.881659   2.233717    -1.74   0.082    -8.259664    .4963463
           2007  |  -4.038947   2.984637    -1.35   0.176    -9.888728    1.810834
           2008  |  -1.922353    2.22307    -0.86   0.387    -6.279491    2.434785
           2009  |  -2.727981   3.160583    -0.86   0.388     -8.92261    3.466648
           2010  |  -2.247629   .1084401   -20.73   0.000    -2.460168    -2.03509
           2011  |  -2.577807   2.134873    -1.21   0.227    -6.762082    1.606467
           2012  |  -2.463136    2.23888    -1.10   0.271     -6.85126    1.924988
           2013  |   -2.24106   2.164606    -1.04   0.301     -6.48361    2.001489
           2014  |  -3.028667   2.189089    -1.38   0.167    -7.319203    1.261869
           2015  |  -3.028177   2.270207    -1.33   0.182    -7.477702    1.421348
           2016  |  -1.851315   2.601384    -0.71   0.477    -6.949935    3.247305
           2017  |  -3.549959       2.21    -1.61   0.108     -7.88148    .7815624
           2018  |  -2.435352   2.251203    -1.08   0.279    -6.847629    1.976925
           2019  |          0    1.74777     0.00   1.000    -3.425567    3.425567
----------------------------------------------------------------------------------

Can you please suggest whether I can go with this model or not since the Z values are negative for some variables and for other it is zero
Thanking you

Last edited by Fadi Ansar; 13 Sep 2021, 10:16.

Comment

Joao Santos Silva

Join Date: Apr 2014

Posts: 2826
#63

13 Sep 2021, 15:14

Dear Fadi Ansar,

It looks as if you answered your own question: it was because of the missing values in the CBT variable that you were getting all those coefficients equal to zero. It is up to you to decide whether the variable should be included in the model, or whether you prefer to exclude it to have a much larger sample.

Best wishes,

Joao
Comment
Fadi Ansar

Join Date: May 2021

Posts: 119
#64

20 Sep 2021, 09:43

Thankyou again Joao Santos Silva .I am again stuck with some problems again ,if possible can you please help me on this
1.so before running quantile regression ,I have plotted quantile plot for one of my dependent variable NPL _asset .I was getting following graph and what does actually it mean as I am not able to interpret this .can I go for a quantile regression for the same variable (I decided for quantile regression because of non normality of data and literature advocated to do so)
Graph _NPL asset(quantile).gph
also attaching the histogram as well
Graph (histogram).gph
2.How can I do F-tests for the equality of quantile slope coefficients for an inter comparison of coefficients -is there any specific command I can use since I couldn't find it anywhere

Thanks in advance .Your help is highly appreciated
Comment
Joao Santos Silva

Join Date: Apr 2014

Posts: 2826
#65

20 Sep 2021, 10:13

1. The plot does not tell me much about your data, but you can use QR as long as your dependent variable is continuous.
2. I am not sure what exactly are the coefficients you want to compare; please clarify.
Comment
Fadi Ansar

Join Date: May 2021

Posts: 119
#66

20 Sep 2021, 12:41

Thankyou @ Joao Santos Silva .I just want to test the equality of coefficients at different quantile (from10 th to 90th) and The t-statistic is for a difference of means test from the first to the 9 th quartile as well,how can i perform both
.with respect to 1 answer ,does my histogram is okay if i want to show the non normality of variable.so there is no interpretation for my quantile plot .
Comment
Joao Santos Silva

Join Date: Apr 2014

Posts: 2826
#67

20 Sep 2021, 12:52

Which command are you using to perform the estimation?
Comment
Fadi Ansar

Join Date: May 2021

Posts: 119
#68

21 Sep 2021, 11:56

Dear Joao Santos Silva ,
I was using sqreg and qregpd command for estimation

Thanks and regards
fadi
Comment
Joao Santos Silva

Join Date: Apr 2014

Posts: 2826
#69

21 Sep 2021, 13:59

I do not know about qregpd; for sqreg you can use standard Stata tests.
Comment
Fadi Ansar

Join Date: May 2021

Posts: 119
#70

22 Sep 2021, 05:21

Thankyou Joao Santos Silva .So for testing the equality of different coefficients ,I can do the normal F test.

Thanks
Fadi
Comment
Joao Santos Silva

Join Date: Apr 2014

Posts: 2826
#71

22 Sep 2021, 06:22

I believe that is the case if you use sqreg.
Comment

Fadi Ansar

Join Date: May 2021
Posts: 119

#72

22 Sep 2021, 09:20

ok .Thanks Joao Santos Silva .I am having one more doubt with respect to qreg command .when adding I. Year variable for running a quantile regression using qreg command ,I was getting blank results -so it is because of adding year effects into the model or is it because qreg command is not supportive of i.variables .so my concerns is whether inclusion dummies is not supported by qreg command
I am attaching both results

Code:

. qreg Zscore d_MP   NIM  lasset CapitalRatio,quantile(.5)
Iteration  1:  WLS sum of weighted deviations =  352.24318

Iteration  1: sum of abs. weighted deviations =  356.74618
Iteration  2: sum of abs. weighted deviations =  352.01914
Iteration  3: sum of abs. weighted deviations =  351.05703
Iteration  4: sum of abs. weighted deviations =  350.69714
Iteration  5: sum of abs. weighted deviations =  350.66093
Iteration  6: sum of abs. weighted deviations =  350.65873
Iteration  7: sum of abs. weighted deviations =  350.65726
Iteration  8: sum of abs. weighted deviations =  350.65616
Iteration  9: sum of abs. weighted deviations =   350.6559

Median regression                                   Number of obs =      1,009
  Raw sum of deviations 360.4623 (about 16.789)
  Min sum of deviations 350.6559                    Pseudo R2     =     0.0272

------------------------------------------------------------------------------
      Zscore |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
        d_MP |   .0535337   .0232238     2.31   0.021     .0079609    .0991065
         NIM |   .0345898   .0288858     1.20   0.231    -.0220937    .0912734
      lasset |    .004897   .0143509     0.34   0.733    -.0232642    .0330582
CapitalRatio |   .0521007   .0165913     3.14   0.002     .0195431    .0846582
       _cons |   16.70362   .2034012    82.12   0.000     16.30448    17.10276
------------------------------------------------------------------------------

.

when adding year effects to the model ,the result is below

Code:

qreg Zscore d_MP   NIM  lasset CapitalRatio i.Year,quantile(.5)
Iteration  1:  WLS sum of weighted deviations =  15.028782

Iteration  1: sum of abs. weighted deviations =   14.42415
Iteration  2: sum of abs. weighted deviations =   14.42415
Iteration  3: sum of abs. weighted deviations =   14.42415
Iteration  4: sum of abs. weighted deviations =   14.42415
Iteration  5: sum of abs. weighted deviations =   14.42415
Iteration  6: sum of abs. weighted deviations =   14.42415
Iteration  7: sum of abs. weighted deviations =   14.42415
Iteration  8: sum of abs. weighted deviations =   14.42415
Iteration  9: sum of abs. weighted deviations =   14.42415
Iteration 10: sum of abs. weighted deviations =   14.42415
note:  alternate solutions exist
Iteration 11: sum of abs. weighted deviations =   14.42415
note:  alternate solutions exist
Iteration 12: sum of abs. weighted deviations =   14.42415
note:  alternate solutions exist
Iteration 13: sum of abs. weighted deviations =   14.42415
Iteration 14: sum of abs. weighted deviations =   14.42415
note:  alternate solutions exist
Iteration 15: sum of abs. weighted deviations =   14.42415
Iteration 16: sum of abs. weighted deviations =   14.42415
Iteration 17: sum of abs. weighted deviations =   14.42415
Iteration 18: sum of abs. weighted deviations =   14.42415
note:  alternate solutions exist
Iteration 19: sum of abs. weighted deviations =   14.42415
Iteration 20: sum of abs. weighted deviations =   14.42415
Iteration 21: sum of abs. weighted deviations =   14.42415
Iteration 22: sum of abs. weighted deviations =   14.42415
Iteration 23: sum of abs. weighted deviations =   14.42415
Iteration 24: sum of abs. weighted deviations =   14.42415
note:  alternate solutions exist
Iteration 25: sum of abs. weighted deviations =   14.42415
Iteration 26: sum of abs. weighted deviations =   14.42415
Iteration 27: sum of abs. weighted deviations =   14.42415
Iteration 28: sum of abs. weighted deviations =   14.42415
Iteration 29: sum of abs. weighted deviations =   14.42415
Iteration 30: sum of abs. weighted deviations =   14.42415
Iteration 31: sum of abs. weighted deviations =   14.42415
Iteration 32: sum of abs. weighted deviations =   14.42415
Iteration 33: sum of abs. weighted deviations =   14.42415
Iteration 34: sum of abs. weighted deviations =   14.42415
Iteration 35: sum of abs. weighted deviations =   14.42415
Iteration 36: sum of abs. weighted deviations =   14.42415
note:  alternate solutions exist
Iteration 37: sum of abs. weighted deviations =   14.42415
note:  alternate solutions exist
Iteration 38: sum of abs. weighted deviations =   14.42415
Iteration 39: sum of abs. weighted deviations =   14.42415
Iteration 40: sum of abs. weighted deviations =   14.42415
Iteration 41: sum of abs. weighted deviations =   14.42415
note:  alternate solutions exist
Iteration 42: sum of abs. weighted deviations =   14.42415
Iteration 43: sum of abs. weighted deviations =   14.42415
note:  alternate solutions exist
Iteration 44: sum of abs. weighted deviations =   14.42415
Iteration 45: sum of abs. weighted deviations =   14.42415
Iteration 46: sum of abs. weighted deviations =   14.42415
Iteration 47: sum of abs. weighted deviations =   14.42415
note:  alternate solutions exist
Iteration 48: sum of abs. weighted deviations =   14.42415
note:  alternate solutions exist
Iteration 49: sum of abs. weighted deviations =   14.42415
Iteration 50: sum of abs. weighted deviations =   14.42415
Iteration 51: sum of abs. weighted deviations =   14.42415
note:  alternate solutions exist
Iteration 52: sum of abs. weighted deviations =   14.42415

Median regression                                   Number of obs =      1,009
  Raw sum of deviations 360.4623 (about 16.789)
  Min sum of deviations 14.42415                    Pseudo R2     =     0.9600

------------------------------------------------------------------------------
      Zscore |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
        d_MP |          0  (omitted)
         NIM |          0  (omitted)
      lasset |   1.37e-22          .        .       .            .           .
CapitalRatio |  -1.91e-17          .        .       .            .           .
             |
        Year |
       2007  |     -.5675          .        .       .            .           .
       2008  |     1.9291          .        .       .            .           .
       2009  |     1.1902          .        .       .            .           .
       2010  |     1.7234          .        .       .            .           .
       2011  |      1.427          .        .       .            .           .
       2012  |      1.561          .        .       .            .           .
       2013  |     1.7067          .        .       .            .           .
       2014  |      .9813          .        .       .            .           .
       2015  |      .8998          .        .       .            .           .
       2016  |     2.0024          .        .       .            .           .
       2017  |      .3995          .        .       .            .           .
       2018  |     1.2323          .        .       .            .           .
       2019  |     3.8993          .        .       .            .           .
             |
       _cons |    15.5567          .        .       .            .           .
------------------------------------------------------------------------------

Thanking you,
Fadi

Comment

Joao Santos Silva

Join Date: Apr 2014

Posts: 2826
#73

22 Sep 2021, 10:04

It looks as if the dummies explain (almost) all the variation in your data. You need to think hard about what you are doing and why you are doing it.
Comment
Fadi Ansar

Join Date: May 2021

Posts: 119
#74

22 Sep 2021, 10:41

Thankyou Joao Santos Silva .I really appreciates your comments .I have included the dummies to capture the year effects and usage of quantile regression is because my dependent variable is not normal and the previous studies similar to this has employed such method. Now I am totally confused whether should i go with out time dummies or change the model
Comment
Joao Santos Silva

Join Date: Apr 2014

Posts: 2826
#75

22 Sep 2021, 14:47

You should discuss this with your supervisor.
Comment

Announcement

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment