Dear
I want to know when and why I can use robust command in Stata for panel regression model?!
I want to know when and why I can use robust command in Stata for panel regression model?!
-
Login or Register
- Log in with
- You are not logged in. You can browse but not post. Login or Register by clicking 'Login or Register' at the top-right of this page. For more information on Statalist, see the FAQ.
-
-
-robust- is an option (not command) that specifies that cluster robust standard errors be reported. In panel data, these are robust to some kinds of misspecification. See
and click on the linked PDF manual entry (and the references therein) for a discussion of this.Code:help xtreg
-
Thank you soo much Dr.AndrewComment
-
Dear Andrew Musau
I have another question I run AMG estimation with and with robust option and then I run the estat vce, corr and after the command with robust the correlation =0 for all my explanatory variables but after the option without robust there are a correlations among variables ? please can you explain why with robust the corr matrix=0??
HTML Code:xtmg dTotalExpenditurePropGDP lx1 lx2 lx3 slopedummy1 slopedummy2 slopedummy3 , aug robust trend Augmented Mean Group estimator (Bond & Eberhardt, 2009; Eberhardt & Teal, 2010) Common dynamic process included as additional regressor All coefficients represent averages across groups (group variable: CountryID) Coefficient averages computed as outlier-robust means (using rreg) Mean Group type estimation Number of obs = 796 AMG Wald chi2(6) = 3.54 Prob > chi2 = 0.7382 ------------------------------------------------------------------------------ dTotalExpe~P | Coefficient Std. err. z P>|z| [95% conf. interval] -------------+---------------------------------------------------------------- lx1 | .0151935 .0193858 0.78 0.433 -.0228019 .053189 lx2 | -.0115215 .0116609 -0.99 0.323 -.0343764 .0113334 lx3 | -.0149404 .0146712 -1.02 0.309 -.0436955 .0138147 slopedummy1 | -.0300774 .0330989 -0.91 0.364 -.09495 .0347953 slopedummy2 | -.0040926 .0149124 -0.27 0.784 -.0333204 .0251352 slopedummy3 | -.0023147 .0192515 -0.12 0.904 -.0400469 .0354174 c_d_p | .4084651 .1071969 3.81 0.000 .198363 .6185671 trend | -.0001642 .0001878 -0.87 0.382 -.0005323 .0002039 _cons | .0011697 .00565 0.21 0.836 -.0099041 .0122435 ------------------------------------------------------------------------------ Root Mean Squared Error (sigma): 0.0564 (RMSE uses residuals from group-specific regressions: unaffected by 'robust'). Variable c_d_p refers to the common dynamic process. Variable trend refers to the group-specific linear trend terms. Share of group-specific trends significant at 5% level: 0.034 (= 1 trends) . estat vce, corr Correlation matrix of coefficients of xtmg model e(V) | lx1 lx2 lx3 sloped~1 sloped~2 sloped~3 c_d_p trend _cons -------------+------------------------------------------------------------------------------------------ lx1 | 1.0000 lx2 | 0.0000 1.0000 lx3 | 0.0000 0.0000 1.0000 slopedummy1 | 0.0000 0.0000 0.0000 1.0000 slopedummy2 | 0.0000 0.0000 0.0000 0.0000 1.0000 slopedummy3 | 0.0000 0.0000 0.0000 0.0000 0.0000 1.0000 c_d_p | 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 1.0000 trend | 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 1.0000 _cons | 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 1.0000
and the following results without robust option
Thanks in AdvanceHTML Code:. xtmg dTotalExpenditurePropGDP lx1 lx2 lx3 slopedummy1 slopedummy2 slopedummy3 , aug trend Augmented Mean Group estimator (Bond & Eberhardt, 2009; Eberhardt & Teal, 2010) Common dynamic process included as additional regressor All coefficients represent averages across groups (group variable: CountryID) Coefficient averages computed as unweighted means Mean Group type estimation Number of obs = 796 AMG Wald chi2(6) = 25.75 Prob > chi2 = 0.0002 ------------------------------------------------------------------------------ dTotalExpe~P | Coefficient Std. err. z P>|z| [95% conf. interval] -------------+---------------------------------------------------------------- lx1 | .0002567 .0280607 0.01 0.993 -.0547412 .0552546 lx2 | -.0333057 .0264025 -1.26 0.207 -.0850536 .0184421 lx3 | -.0371906 .0244625 -1.52 0.128 -.0851362 .0107551 slopedummy1 | -.0028595 .0525642 -0.05 0.957 -.1058835 .1001644 slopedummy2 | .0082383 .0379022 0.22 0.828 -.0660487 .0825252 slopedummy3 | .0056132 .0352261 0.16 0.873 -.0634287 .0746551 c_d_p | .9166282 .3837471 2.39 0.017 .1644977 1.668759 trend | .0001598 .0005796 0.28 0.783 -.0009762 .0012958 _cons | -.0140172 .0244095 -0.57 0.566 -.0618589 .0338245 ------------------------------------------------------------------------------ Root Mean Squared Error (sigma): 0.0564 Variable c_d_p refers to the common dynamic process. Variable trend refers to the group-specific linear trend terms. Share of group-specific trends significant at 5% level: 0.034 (= 1 trends) . estat vce, corr Correlation matrix of coefficients of xtmg model e(V) | lx1 lx2 lx3 sloped~1 sloped~2 sloped~3 c_d_p trend _cons -------------+------------------------------------------------------------------------------------------ lx1 | 1.0000 lx2 | 0.0051 1.0000 lx3 | 0.6117 -0.4777 1.0000 slopedummy1 | -0.9330 -0.0803 -0.5047 1.0000 slopedummy2 | 0.2486 -0.9171 0.6065 -0.2269 1.0000 slopedummy3 | -0.2022 0.6431 -0.8070 0.0423 -0.5423 1.0000 c_d_p | -0.7436 0.0499 -0.6813 0.7451 -0.4010 0.1653 1.0000 trend | -0.2870 -0.0004 -0.5359 0.1493 0.0901 0.6254 0.0898 1.0000 _cons | 0.4319 0.0643 0.6144 -0.3100 -0.0874 -0.6106 -0.2458 -0.9790 1.0000 .
Comment
-
xtmg is from SSC. Read the documentation:
As the command computes no covariances if specifying the option -robust-, the off-diagonal elements of the VCE matrix are empty.robust estimates the outlier-robust mean of parameter coefficients across groups. This is implemented via the Stata command rreg for robust regression. An example
of this practice can be found in Bond, Leblebicioglu and Schiantarelli (2010). This option is not to be confused with the standard option calling for White
heteroskedasticity-robust standard errors in the reg and xtreg commands.
Comment
Comment