* Panel Data Models in Stata

* Copyright 2013 by Ani Katchova

clear all
set more off
use C:\Econometrics\Data\panel_wage
global
global
global
global

id id
t t
ylist lwage
xlist exp exp2 wks ed

describe $id $t $ylist $xlist

summarize $id $t $ylist $xlist
* Set data as panel data
sort $id $t
xtset $id $t
xtdescribe
xtsum $id $t $ylist $xlist
* Pooled OLS estimator
reg $ylist $xlist
* Population-averaged estimator
xtreg $ylist $xlist, pa
* Between estimator
xtreg $ylist $xlist, be
* Fixed effects or within estimator
xtreg $ylist $xlist, fe
* First-differences estimator
reg D.($ylist $xlist), noconstant
* Random effects estimator
xtreg $ylist $xlist, re theta
* Hausman test for fixed versus random effects model
quietly xtreg $ylist $xlist, fe
estimates store fixed
quietly xtreg $ylist $xlist, re
estimates store random
hausman fixed random
* Breusch-Pagan LM test for random effects versus OLS
quietly xtreg $ylist $xlist, re
xttest0
* Recovering individual-specific effects
quietly xtreg $ylist $xlist, fe
predict alphafehat, u
sum alphafehat

. * Panel Data Models in Stata

. * Copyright 2013 by Ani Katchova
.
. clear all
. set more off
.
. use C:\Econometrics\Data\panel_wage
(PSID wage data 1976-82 from Baltagi and Khanti-Akom (1990))
.
. global id id
. global t t
. global ylist lwage
. global xlist exp exp2 wks ed
.
. describe $id $t $ylist $xlist
storage display
value
variable name
type
format
label
variable label
-------------------------------------------------------------------------------------------------------id
float %9.0g
t
float %9.0g
lwage
float %9.0g
log wage
exp
float %9.0g
years of full-time work experience
exp2
float %9.0g
wks
float %9.0g
weeks worked
ed
float %9.0g
years of education
. summarize $id $t $ylist $xlist
Variable |
Obs
Mean
Std. Dev.
Min
Max
-------------+-------------------------------------------------------id |
4165
298
171.7821
1
595
t |
4165
4
2.00024
1
7
lwage |
4165
6.676346
.4615122
4.60517
8.537
exp |
4165
19.85378
10.96637
1
51
exp2 |
4165
514.405
496.9962
1
2601
-------------+-------------------------------------------------------wks |
4165
46.81152
5.129098
5
52
ed |
4165
12.84538
2.787995
4
17
.
. * Set data as panel data
. sort $id $t
. xtset $id $t
panel variable:

id (strongly balanced)

time variable:
delta:

t, 1 to 7
1 unit

. xtdescribe
id:
t:

1, 2, ..., 595
1, 2, ..., 7
Delta(t) = 1 unit
Span(t) = 7 periods
(id*t uniquely identifies each observation)

Distribution of T_i:

min
7

5%
7

25%
7

50%
7

n =
T =

75%
7

595
7

95%
7

max
7

Freq. Percent
Cum. | Pattern
---------------------------+--------595
100.00 100.00 | 1111111
---------------------------+--------595
100.00
| XXXXXXX
. xtsum $id $t $ylist $xlist
Variable
|
Mean
Std. Dev.
Min
Max |
Observations
-----------------+--------------------------------------------+---------------id
overall |
298
171.7821
1
595 |
N =
4165
between |
171.906
1
595 |
n =
595
within |
0
298
298 |
T =
7
|
|
t
overall |
4
2.00024
1
7 |
N =
4165
between |
0
4
4 |
n =
595
within |
2.00024
1
7 |
T =
7
|
|
lwage
overall | 6.676346
.4615122
4.60517
8.537 |
N =
4165
between |
.3942387
5.3364
7.813596 |
n =
595
within |
.2404023
4.781808
8.621092 |
T =
7
|
|
exp
overall | 19.85378
10.96637
1
51 |
N =
4165
between |
10.79018
4
48 |
n =
595
within |
2.00024
16.85378
22.85378 |
T =
7
|
|
exp2
overall |
514.405
496.9962
1
2601 |
N =
4165
between |
489.0495
20
2308 |
n =
595
within |
90.44581
231.405
807.405 |
T =
7
|
|
wks
overall | 46.81152
5.129098
5
52 |
N =
4165
between |
3.284016
31.57143
51.57143 |
n =
595
within |
3.941881
12.2401
63.66867 |
T =
7
|
|
ed
overall | 12.84538
2.787995
4
17 |
N =
4165
between |
2.790006
4
17 |
n =
595
within |
0
12.84538
12.84538 |
T =
7
.
. * Pooled OLS estimator
. reg $ylist $xlist

Source |
SS
df
MS
-------------+-----------------------------Model | 251.491445
4 62.8728613
Residual | 635.413457 4160 .152743619
-------------+-----------------------------Total | 886.904902 4164 .212993492

Number of obs
F( 4, 4160)
Prob > F
R-squared
Adj R-squared
Root MSE

=
=
=
=
=
=

4165
411.62
0.0000
0.2836
0.2829
.39082

-----------------------------------------------------------------------------lwage |
Coef.
Std. Err.
t
P>|t|
[95% Conf. Interval]
-------------+---------------------------------------------------------------exp |
.044675
.0023929
18.67
0.000
.0399838
.0493663
exp2 | -.0007156
.0000528
-13.56
0.000
-.0008191
-.0006121
wks |
.005827
.0011827
4.93
0.000
.0035084
.0081456
ed |
.0760407
.0022266
34.15
0.000
.0716754
.080406
_cons |
4.907961
.0673297
72.89
0.000
4.775959
5.039963
-----------------------------------------------------------------------------.
. * Population-averaged estimator
. xtreg $ylist $xlist, pa
Iteration
Iteration
Iteration
Iteration
Iteration
Iteration
Iteration
Iteration
Iteration
Iteration
Iteration
Iteration

1: tolerance = .10073745
2: tolerance = .10171516
3: tolerance = .0939929
4: tolerance = .05375245
5: tolerance = .01851339
6: tolerance = .00488708
7: tolerance = .00118383
8: tolerance = .00028047
9: tolerance = .0000661
10: tolerance = .00001556
11: tolerance = 3.660e-06
12: tolerance = 8.612e-07

GEE population-averaged model

Group variable:
id
Link:
identity
Family:
Gaussian
Correlation:
exchangeable
Scale parameter:

.7476287

Number of obs
Number of groups
Obs per group: min
avg
max
Wald chi2(4)
Prob > chi2

=
=
=
=
=
=
=

4165
595
7
7.0
7
6160.57
0.0000

-----------------------------------------------------------------------------lwage |
Coef.
Std. Err.
z
P>|z|
[95% Conf. Interval]
-------------+---------------------------------------------------------------exp |
.1079955
.0024527
44.03
0.000
.1031883
.1128026
exp2 | -.0005202
.0000543
-9.59
0.000
-.0006266
-.0004139
wks |
.0008365
.0006042
1.38
0.166
-.0003477
.0020208
ed |
.1378558
.0125814
10.96
0.000
.1131968
.1625149
_cons |
2.98986
.1711799
17.47
0.000
2.654353
3.325366
-----------------------------------------------------------------------------.
. * Between estimator

. xtreg $ylist $xlist, be

Between regression (regression on group means)
Group variable: id

Number of obs
Number of groups

=
=

4165
595

R-sq:

Obs per group: min =

avg =
max =

7
7.0
7

within = 0.1357
between = 0.3264
overall = 0.2723

sd(u_i + avg(e_i.))=

.324656

F(4,590)
Prob > F

=
=

71.48
0.0000

-----------------------------------------------------------------------------lwage |
Coef.
Std. Err.
t
P>|t|
[95% Conf. Interval]
-------------+---------------------------------------------------------------exp |
.038153
.0056967
6.70
0.000
.0269647
.0493412
exp2 | -.0006313
.0001257
-5.02
0.000
-.0008781
-.0003844
wks |
.0130903
.0040659
3.22
0.001
.0051048
.0210757
ed |
.0737838
.0048985
15.06
0.000
.0641632
.0834044
_cons |
4.683039
.2100989
22.29
0.000
4.270407
5.095672
-----------------------------------------------------------------------------.
. * Fixed effects or within estimator
. xtreg $ylist $xlist, fe
note: ed omitted because of collinearity
Fixed-effects (within) regression
Group variable: id

Number of obs
Number of groups

=
=

4165
595

R-sq:

Obs per group: min =

avg =
max =

7
7.0
7

within = 0.6566
between = 0.0276
overall = 0.0476

corr(u_i, Xb)

= -0.9107

F(3,3567)
Prob > F

=
=

2273.74
0.0000

-----------------------------------------------------------------------------lwage |
Coef.
Std. Err.
t
P>|t|
[95% Conf. Interval]
-------------+---------------------------------------------------------------exp |
.1137879
.0024689
46.09
0.000
.1089473
.1186284
exp2 | -.0004244
.0000546
-7.77
0.000
-.0005315
-.0003173
wks |
.0008359
.0005997
1.39
0.163
-.0003399
.0020116
ed |
0 (omitted)
_cons |
4.596396
.0389061
118.14
0.000
4.520116
4.672677
-------------+---------------------------------------------------------------sigma_u | 1.0362039
sigma_e | .15220316
rho | .97888036
(fraction of variance due to u_i)
-----------------------------------------------------------------------------F test that all u_i=0:
F(594, 3567) =
56.52
Prob > F = 0.0000
.
. * First-differences estimator
. reg D.($ylist $xlist), noconstant

note: _delete omitted because of collinearity

Source |
SS
df
MS
-------------+-----------------------------Model | 33.3371458
3 11.1123819
Residual |
117.57812 3567 .032962748
-------------+-----------------------------Total | 150.915266 3570 .042273184

Number of obs
F( 3, 3567)
Prob > F
R-squared
Adj R-squared
Root MSE

=
=
=
=
=
=

3570
337.12
0.0000
0.2209
0.2202
.18156

-----------------------------------------------------------------------------D.lwage |
Coef.
Std. Err.
t
P>|t|
[95% Conf. Interval]
-------------+---------------------------------------------------------------exp |
D1. |
.1170654
.0063106
18.55
0.000
.1046927
.1294381
|
exp2 |
D1. | -.0005321
.0001393
-3.82
0.000
-.0008052
-.0002591
|
wks |
D1. | -.0002683
.0005648
-0.47
0.635
-.0013757
.0008392
|
ed |
D1. |
0 (omitted)
-----------------------------------------------------------------------------.
. * Random effects estimator
. xtreg $ylist $xlist, re theta
Random-effects GLS regression
Group variable: id

Number of obs
Number of groups

=
=

4165
595

R-sq:

Obs per group: min =

avg =
max =

7
7.0
7

within = 0.6340
between = 0.1716
overall = 0.1830

corr(u_i, X)
theta

= 0 (assumed)
= .82280511

Wald chi2(4)
Prob > chi2

=
=

3012.45
0.0000

-----------------------------------------------------------------------------lwage |
Coef.
Std. Err.
z
P>|z|
[95% Conf. Interval]
-------------+---------------------------------------------------------------exp |
.0888609
.0028178
31.54
0.000
.0833382
.0943837
exp2 | -.0007726
.0000623
-12.41
0.000
-.0008946
-.0006505
wks |
.0009658
.0007433
1.30
0.194
-.000491
.0024226
ed |
.1117099
.0060572
18.44
0.000
.0998381
.1235818
_cons |
3.829366
.0936336
40.90
0.000
3.645848
4.012885
-------------+---------------------------------------------------------------sigma_u | .31951859
sigma_e | .15220316
rho | .81505521
(fraction of variance due to u_i)
-----------------------------------------------------------------------------.

. * Hausman test for fixed versus random effects model

. quietly xtreg $ylist $xlist, fe
. estimates store fixed
. quietly xtreg $ylist $xlist, re
. estimates store random
. hausman fixed random
---- Coefficients ---|
(b)
(B)
(b-B)
sqrt(diag(V_b-V_B))
|
fixed
random
Difference
S.E.
-------------+---------------------------------------------------------------exp |
.1137879
.0888609
.0249269
.
exp2 |
-.0004244
-.0007726
.0003482
.
wks |
.0008359
.0009658
-.0001299
.
-----------------------------------------------------------------------------b = consistent under Ho and Ha; obtained from xtreg
B = inconsistent under Ha, efficient under Ho; obtained from xtreg
Test:

Ho:

difference in coefficients not systematic

chi2(3) = (b-B)'[(V_b-V_B)^(-1)](b-B)
=
6191.43
Prob>chi2 =
0.0000
(V_b-V_B is not positive definite)

.
. * Breusch-Pagan LM test for random effects versus OLS
. quietly xtreg $ylist $xlist, re
. xttest0
Breusch and Pagan Lagrangian multiplier test for random effects
lwage[id,t] = Xb + u[id] + e[id,t]
Estimated results:
|
Var
sd = sqrt(Var)
---------+----------------------------lwage |
.2129935
.4615122
e |
.0231658
.1522032
u |
.1020921
.3195186
Test:

Var(u) = 0
chibar2(01) =
Prob > chibar2 =

.
. * Recovering individual-specific effects
. quietly xtreg $ylist $xlist, fe
. predict alphafehat, u

5192.13
0.0000

. sum alphafehat
Variable |
Obs
Mean
Std. Dev.
Min
Max
-------------+-------------------------------------------------------alphafehat |
4165
-1.97e-10
1.035457 -3.700898
1.896135