Sie sind auf Seite 1von 9

Course:

Econometrics (Course Number 172843/76443)

Date and Time:

29th of July 2014; 14:45 - 15:45

Please answer all questions. The maximum credits achievable for each question is
indicated in brackets.
Question 1:

(30)

Have a look at the following STATA output. These are results from a study on the
determinants of wages.
lwage:
hours:
age:
educ:
tenure:
exper:
exper2:
black:
union:

logarithm of hourly wage


usual hours worked per week
age of individual
number of completed years of education
number of years an individual has worked in the present company
work experience
square of work experience
dummy=1 if individual afro-american
dummy=1 if individual member of a labor union

. xtreg lwage hours age educ tenure exper exper2 black union, fe
note: educ omitted because of collinearity
note: black omitted because of collinearity
Fixed-effects (within) regression
Group variable: id

Number of obs
Number of groups

=
=

3580
716

R-sq:

Obs per group: min =


avg =
max =

5
5.0
5

within = 0.1610
between = 0.0960
overall = 0.1067

corr(u_i, Xb)

= -0.0044

F(6,2858)
Prob > F

=
=

91.39
0.0000

-----------------------------------------------------------------------------lwage |
Coef.
Std. Err.
t
P>|t|
[95% Conf. Interval]
-------------+---------------------------------------------------------------hours | -.0065762
.0007367
-8.93
0.000
-.0080207
-.0051318
age | -.0095017
.0090391
-1.05
0.293
-.0272255
.008222
educ |
0 (omitted)
tenure |
.0012119
.0017399
0.70
0.486
-.0021997
.0046235
exper |
.0690974
.0119522
5.78
0.000
.0456616
.0925333
exper2 | -.0011863
.0002277
-5.21
0.000
-.0016327
-.0007399
black |
0 (omitted)
union |
.0742261
.0141297
5.25
0.000
.0465207
.1019315
_cons |
1.83943
.2103743
8.74
0.000
1.42693
2.251931
-------------+---------------------------------------------------------------sigma_u | .40405192
sigma_e | .19305019
rho | .81414721
(fraction of variance due to u_i)
-----------------------------------------------------------------------------F test that all u_i=0:
F(715, 2858) =
21.27
Prob > F = 0.0000

a) Interpret the output completely, including naming the model used, an economic
interpretation of all coefficients and an interpretation of the econometric tests and the key
figures STATA displays.

b) Compare the above output to the following STATA outputs. Which model do you
prefer and why?
. xtreg lwage hours age educ tenure exper exper2 black union, re
Random-effects GLS regression
Group variable: id

Number of obs
Number of groups

=
=

3580
716

R-sq:

Obs per group: min =


avg =
max =

5
5.0
5

within = 0.1598
between = 0.3368
overall = 0.3077

corr(u_i, X)

= 0 (assumed)

Wald chi2(8)
Prob > chi2

=
=

909.37
0.0000

-----------------------------------------------------------------------------lwage |
Coef.
Std. Err.
z
P>|z|
[95% Conf. Interval]
-------------+---------------------------------------------------------------hours | -.0057205
.0006928
-8.26
0.000
-.0070783
-.0043626
age | -.0076863
.003088
-2.49
0.013
-.0137386
-.001634
educ |
.0749508
.0053806
13.93
0.000
.0644051
.0854966
tenure |
.0031691
.0015491
2.05
0.041
.0001329
.0062053
exper |
.0653072
.0063681
10.26
0.000
.052826
.0777885
exper2 | -.0011775
.0002185
-5.39
0.000
-.0016057
-.0007493
black | -.1412562
.0291537
-4.85
0.000
-.1983965
-.084116
union |
.0898778
.013117
6.85
0.000
.0641689
.1155867
_cons |
.8320177
.1153578
7.21
0.000
.6059207
1.058115
-------------+---------------------------------------------------------------sigma_u | .33177812
sigma_e | .19305019
rho | .74706741
(fraction of variance due to u_i)
-----------------------------------------------------------------------------. estimates store re
. hausman fe re
---- Coefficients ---|
(b)
(B)
(b-B)
sqrt(diag(V_b-V_B))
|
fe
re
Difference
S.E.
-------------+---------------------------------------------------------------hours |
-.0065762
-.0057205
-.0008558
.0002504
age |
-.0095017
-.0076863
-.0018154
.0084952
tenure |
.0012119
.0031691
-.0019572
.0007922
exper |
.0690974
.0653072
.0037902
.0101145
exper2 |
-.0011863
-.0011775
-8.77e-06
.000064
union |
.0742261
.0898778
-.0156517
.0052528
-----------------------------------------------------------------------------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(6) = (b-B)'[(V_b-V_B)^(-1)](b-B)
=
26.53
Prob>chi2 =
0.0002

. 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 |
.2158595
.4646068
e |
.0372684
.1930502
u |
.1100767
.3317781
Test:

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

3912.32
0.0000

c) Describe and explain the difference between random effects and fixed effects.

Question 2:

(8)

Consider the following Stata command:


reg depvar var1 var2 c.var1#c.var2
(Note: depvar is a dependent variable, var1 and var2 are two independent variables)
What is included in the regression here? How would you interpret var1 and var2?

Question 3:

(12)

How can a Hausman test show that endogeneity is a problem? Please explain the steps
briefly and include the involved variables in your explanation.

Question 4:

(10)

a) Describe two time series models you know and name at least two different
questions which can be answered with these models.

b) Calculate the multiplier effect of totr from the following output (round each
value to the second digit):
----------------------------------------------------------lquan | Coef.
Std. Err.
t
P>|t|
-------------+--------------------------------------------lprice | -.1853887
.3056501
-0.61
0.549
|
totr |
--. |
46.30765
17.91052
2.59
0.015
L1. | -7.663706
22.8905
-0.33
0.740
L2. | -4.418813
24.27131
-0.18
0.857
L3. |
16.00065
15.94317
1.00
0.324
|
rainy | -.0782845
.2126974
-0.37
0.716
mixed |
.0232826
.232216
0.10
0.921
stormy| -.3232928
.2627149
-1.23
0.229
_cons |
8.463075
.2738453
30.90
0.000
-----------------------------------------------------------

Das könnte Ihnen auch gefallen