[Limdep Nlogit List] random effects vs OLS
Andreas Drichoutis
adrihout at aua.gr
Fri Jan 15 04:46:43 EST 2010
Dear all,
Is it correct that I'm getting the exact same coefficient estimates from a
random effects model with OLS?
Yours,
Andreas Drichoutis
--> skip $
--> regress ; lhs=bet2lot6 ; rhs=x1 ; pds=10; panel ; random $
+----------------------------------------------------+
| OLS Without Group Dummy Variables |
| Ordinary least squares regression |
| Model was estimated Jan 14, 2010 at 07:41:04PM |
| LHS=BET2LOT6 Mean = 2.474676 |
| Standard deviation = 2.444271 |
| WTS=none Number of observs. = 710 |
| Model size Parameters = 12 |
| Degrees of freedom = 698 |
| Residuals Sum of squares = 3318.380 |
| Standard error of e = 2.180396 |
| Fit R-squared = .2166045 |
| Adjusted R-squared = .2042588 |
| Model test F[ 11, 698] (prob) = 17.54 (.0000) |
| Diagnostic Log likelihood = -1554.845 |
| Restricted(b=0) = -1641.506 |
| Chi-sq [ 11] (prob) = 173.32 (.0000) |
| Info criter. LogAmemiya Prd. Crt. = 1.575773 |
| Akaike Info. Criter. = 1.575770 |
| Bayes Info. Criter. = 1.652929 |
+----------------------------------------------------+
+----------------------------------------------------+
| Panel Data Analysis of BET2LOT6 [ONE way] |
| Unconditional ANOVA (No regressors) |
| Source Variation Deg. Free. Mean Square |
| Between 3030.11 70. 43.2872 |
| Residual 1205.79 639. 1.88699 |
| Total 4235.89 709. 5.97446 |
+----------------------------------------------------+
+--------+--------------+----------------+--------+--------+----------+
|Variable| Coefficient | Standard Error |b/St.Er.|P[|Z|>z]| Mean of X|
+--------+--------------+----------------+--------+--------+----------+
|GENDER | 1.44023*** .18075193 7.968 .0000 .3943662|
|AGE | .33061*** .05252022 6.295 .0000 20.732394|
|HSIZE | .12633 .08126361 1.555 .1200 4.3802817|
|TOTFEE | .17489* .10456907 1.672 .0944 16.711268|
|ECONPOS1| 2.08826*** .41715206 5.006 .0000 .2676056|
|ECONPOS2| 2.13156*** .41055413 5.192 .0000 .2676056|
|ECONPOS3| 1.99186*** .39024486 5.104 .0000 .3239437|
|ECONPOS4| 1.20966** .47230779 2.561 .0104 .0845070|
|PERIOD | .20266*** .02848913 7.114 .0000 5.5000000|
|TREAT_PR| .60233*** .17831542 3.378 .0007 .4929577|
|TREAT_TR| .36935** .17360452 2.128 .0334 .5070423|
|Constant| -11.8993*** 2.17624091 -5.468 .0000 |
+--------+------------------------------------------------------------+
| Note: ***, **, * = Significance at 1%, 5%, 10% level. |
+---------------------------------------------------------------------+
+----------------------------------------------------+
| Panel:Groups Empty 0, Valid data 71 |
| Smallest 10, Largest 10 |
| Average group size 10.00 |
| There are 10 vars. with no within group variation. |
| GENDER AGE HSIZE TOTFEE ECONPOS1 more |
+----------------------------------------------------+
+--------------------------------------------------+
| Random Effects Model: v(i,t) = e(i,t) + u(i) |
| Estimates: Var[e] = .153695D+01 |
| Var[u] = .321717D+01 |
| Corr[v(i,t),v(i,s)] = .676712 |
| Lagrange Multiplier Test vs. Model (3) = 1463.56 |
| ( 1 df, prob value = .000000) |
| (High values of LM favor FEM/REM over CR model.) |
| Baltagi-Li form of LM Statistic = 1463.56 |
| Sum of Squares .331838D+04 |
| R-squared .216605D+00 |
+--------------------------------------------------+
+--------+--------------+----------------+--------+--------+----------+
|Variable| Coefficient | Standard Error |b/St.Er.|P[|Z|>z]| Mean of X|
+--------+--------------+----------------+--------+--------+----------+
|GENDER | 1.44023*** .48130291 2.992 .0028 .3943662|
|AGE | .33061** .13984988 2.364 .0181 20.732394|
|HSIZE | .12633 .21638724 .584 .5593 4.3802817|
|TOTFEE | .17489 .27844458 .628 .5299 16.711268|
|ECONPOS1| 2.08826* 1.11078482 1.880 .0601 .2676056|
|ECONPOS2| 2.13156* 1.09321596 1.950 .0512 .2676056|
|ECONPOS3| 1.99186* 1.03913680 1.917 .0553 .3239437|
|ECONPOS4| 1.20966 1.25765247 .962 .3361 .0845070|
|PERIOD | .20266*** .01619848 12.511 .0000 5.5000000|
|TREAT_PR| .60233 .47481500 1.269 .2046 .4929577|
|TREAT_TR| .36935 .46227091 .799 .4243 .5070423|
|Constant| -11.8993** 5.78050082 -2.059 .0395 |
+--------+------------------------------------------------------------+
| Note: ***, **, * = Significance at 1%, 5%, 10% level. |
+---------------------------------------------------------------------+
More information about the Limdep
mailing list