Results of random effects in stata

[Anuradha Saikia]

Hello everyone.

Hello everyone. I got the following result for random effects model in stata. The p -value for f statistic does not look great. However Hausman test suggests for a RE model. what should i do.

[Carlo Lazzaro]

Anuradha:
1) robust-clustered standard errors are recommended (149 panels!);
2) the R-sq between is poor;
3) what is the -xttest0- outcome?
4) did you test the model for possible misspecification of the functional form of the regressand?

[Anuradha Saikia]

How to test for misspecification? xttest0 shows random effects is recommended. also firm size having a negative relationship with ROA seems wrong

[Carlo Lazzaro]

Anuradha:
1) run an auxiliry regression as follows:

xtreg ROA fitted_values fitted_values^2, vce(cluster panelid)

If the fitted_sq turn out to be significant, the model is misspecified;
2) the coefficient you'rereferring to does not reach statistical significance: hence, do not worry, as it show no evidence of a negative correlation.

[Anuradha Saikia]

Thank you Carlo Lazzaro

[Anuradha Saikia]

Thank you Carlo Lazzaro .

[Anuradha Saikia]

The fitted values refer to the predictors ?

[Carlo Lazzaro]

Anuradha:
please see the following toy-example of a deliberately misspecified -xtreg,re-:

. use "https://www.stata-press.com/data/r17/nlswork.dta"
(National Longitudinal Survey of Young Women, 14-24 years old in 1968)

. xtreg ln_wage c.age##c.age, vce(cluster idcode)

Random-effects GLS regression Number of obs = 28,510
Group variable: idcode Number of groups = 4,710

R-squared: Obs per group:
Within = 0.1087 min = 1
Between = 0.1015 avg = 6.1
Overall = 0.0870 max = 15

Wald chi2(2) = 1258.33
corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000

(Std. err. adjusted for 4,710 clusters in idcode)
------------------------------------------------------------------------------
| Robust
ln_wage | Coefficient std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
age | .0590339 .0041049 14.38 0.000 .0509884 .0670795
|
c.age#c.age | -.0006758 .0000688 -9.83 0.000 -.0008107 -.000541
|
_cons | .5479714 .0587198 9.33 0.000 .4328826 .6630601
-------------+----------------------------------------------------------------
sigma_u | .3654049
sigma_e | .30245467
rho | .59342665 (fraction of variance due to u_i)
------------------------------------------------------------------------------

. predict fitted, xb
(24 missing values generated)

. g sq_fitted=fitted^2
(24 missing values generated)

. xtreg ln_wage fitted sq_fitted , vce(cluster idcode)

Random-effects GLS regression Number of obs = 28,510
Group variable: idcode Number of groups = 4,710

R-squared: Obs per group:
Within = 0.1088 min = 1
Between = 0.1045 avg = 6.1
Overall = 0.0887 max = 15

Wald chi2(2) = 1316.74
corr(u_i, X) = 0 (assumed) Prob > chi2 = 0.0000

(Std. err. adjusted for 4,710 clusters in idcode)
------------------------------------------------------------------------------
| Robust
ln_wage | Coefficient std. err. z P>|z| [95% conf. interval]
-------------+----------------------------------------------------------------
fitted | 2.805959 .6246598 4.49 0.000 1.581648 4.030269
sq_fitted | -.5516341 .1920793 -2.87 0.004 -.9281026 -.1751656
_cons | -1.468083 .5055433 -2.90 0.004 -2.45893 -.4772365
-------------+----------------------------------------------------------------
sigma_u | .36481589
sigma_e | .30242516
rho | .59269507 (fraction of variance due to u_i)
------------------------------------------------------------------------------

.

[Anuradha Saikia]

Thank you Carlo Lazzaro . I found 649 mi

ssing values. and this is the output

[Carlo Lazzaro]

Anuradha:
the outcome tells you that your model suffers from mispsecification. Therefore, you have to add more predictors and/or interactions in the right-hand side of your regression equation.

[Anuradha Saikia]

Thank you @CarloLazzaro. I have not run the full model yet. The other variables I have are interaction variables. I will run the complete model and see. Thanks again.These things are so easily missed and not taken care of.