Você está na página 1de 7

Econometrics 206-1

Exam III: 10.10 AM -11.40 AM, 24 April 2017

In answering these below, paste the Stata output only when it is asked. When
pasting output, use the copy as picture option. When testing a hypothesis, be sure
to mention the distribution of the test statistic, its degrees of freedom, the level of
significance and the associated critical value. DO NOT USE THE STATA test
COMMAND.

It would be easiest if you inserted your answer between the questions below and
returned this document. Rename the document as `your name.docx’ and upload it
on LMS.

You have to do this exam by yourself. You are allowed to consult the textbook and
your notes. You are NOT allowed to consult anybody whether by speaking, by text
messages or email or any other means. Violations will attract penalties as per
Ashoka policy.

1. (a) Regress log of wages on a constant and the female dummy. Paste output
here.

The above shows the regression of log of wages on the female dummy variable.

lwages(hat)= 4.123 -0.855female

Here, female=1 if individual is a female & female=0 if individual is a male. The


sample contains 1000 observations (N=1000), R^2=0.1537, and the regression
shows that the intercept (beta0 hat)= 4.123, which is the average log(wage) for
men, and the coefficient on the female dummy variable (delta0 hat)= -0.855,
which is the difference in the average wage between women and men.

(b) Interpret the coefficient on the female dummy.

The coefficient on the female dummy is -0.855. It shows the percentage


difference between the earnings of men and women. This implies that women
earn about (100)*(0.855)=85.5% less than men. It shows the discrimination
against women in regard to wage earnings.
(c) Test the null hypothesis that the coefficient on female dummy is -0.5 against
the alternative that the coefficient on female dummy is less than -0.5. Show your
workings.
[5+5+10]

As mentioned in (a), we take the coefficient on female dummy to be delta0.

Null hypothesis- H0: delta0=-0.5

Alternative hypothesis- HA: delta0< -0.5

To test the null hypothesis, we use the T statistic. The degree of freedom= n-k-1=
1000- 1- 1= 998, where n is the number of observations and k is the number of
independent variables. Since the df>120, the distribution is normally distributed.

We have, tfemale= [-0.855 – (-0.5)]/0.064 = -0.355/0.064 = -5.547

We check this at various significance levels. At significance level of 5%, c (critical


value)= -1.645.

In order to reject the null, tfemale< -c. Here, -5.547 < -1.645. therefore, we reject
the null and conclude that female is statistically significant at the 5% significance
level.

At the significance level of 10%, c= -1.282. Again, t< -c. Therefore, we reject the
null and see that female is statistically significant at the 1% level.

2. (a) Regress log of wages on a constant, the female dummy, age of the
individual and the square of age. Paste your output here.

We generate a variable for square of age of individual and this is call sqage.

The above shows the regression of log of wages on a constant, female dummy
variable, age, sqage. N=1000 and R2=0.2236.

(b) Controlling for age and the square of age does not seem to substantially
change the coefficient of the female dummy. Why is that so?
The coefficient on the female dummy variable is -0.858. This shows the
percentage difference in the wage earnings of men and women, controlling for
other factors like age of the individual. It implies that women earn about
(100)*(0.858)= 85.8% less as compared to men, when age has been controlled
for, i.e., given the same age of the individual, women earn about 85.8% less than
men. This is not very different from the result we obtained in (i), when age had
not been controlled for. The difference is only 0.3 (0.35%) which is very
insignificant. This is because age as a variable does not affect wage earnings
much.

[5+5]

3. (a) Regress log of wages on a constant, the female dummy, age of the
individual the square of age and the social group dummies for scheduled caste,
for scheduled tribe and for other backward caste. Note the omitted category is
the general castes (or forward castes). Paste your output here.

(b) Test the null hypothesis that none of the social group dummmies matter, i.e.,
controlling for sex, age and square of age, the average of log wages is the same
for all categories: scheduled castes, scheduled tribes, other backward castes and
the general (forward) castes. Do NOT use the Stata test command.

H0: beta3=beta4=beta5=0
HA: H0 is not true.

Here to test the above hypothesis, we find the F stat.The unrestricted model is
the one which includes the dummy female variable, age, sqage, scd, std, obc- the
regression of which is shown in (a). the restricted model is the one with
variables female, age and sqage- the regression is shown in 2(a). The numerator
df is q=3, where q is number of dropped variables and denominator df is n-k-1=
1000-6-1=993.

F = [(0.2541- 0.2236)/(1 – 0.2541)]*[993/3] = 13.535

At the 5% significance level, c= 2.60. Since F>c, we reject the null.


At 10% level, c=2.08. tehrefore, we reject the null.
At the1% level, c= 2.08. therefoe, reject the null.

(c) Test the null hypothesis that relative to the general (forward) castes,
scheduled castes and other backward castes suffer the same extent of
discrimination. If this requires new regressions, paste the output in your
answer.

[5+15+15]

Let theta1= beta3=beta5, which is beta3-beta5=0, theta1=0


Beta3= theta1 + beta5
H0: theta1=0
HA: H0 is not true.

ttheta1= -0.06/.067= -0.896

degree of freedom= n-k-1= 1000-6-1=993

Testing for a two-tailed hypothesis at the 5% significance level, c= 1.96.

To reject the null, |t|>c.

Here, 0.896<1.96, therefore fails to reject the null at the 5% level.


Even at the 1% and 10% significance levels, the null is not rejected.
4. (a) Regress log of wages on a constant, the female dummy, age of the
individual the square of age, the social group dummies for scheduled caste, for
scheduled tribe and for other backward caste, and the education dummies for
illiterate, literate, primary, secondary, and higher secondary. Paste the output
here.

(b) Compare the above regression with the regression in question 3 (without the
education dummies). Does the inclusion of education dummies alter the
discrimination against women, scheduled castes, scheduled tribes and other
backward castes? Why?

For women-

Coefficient in unrestricted model= -0.63


Coefficient in restricted model= -0.81

Yes. It reduces the discrimination against women by 0.18.

For scheduled castes-

Coefficient in unrestricted model= -0.23


Coefficient in restricted model= -0.44
Yes. Inclusion of education dummies reduces discrimination against scheduled
castes by 0.21

Likewise, for schedules tribes, it reduces discrimination by 0.155 and for obcs, it
reduces by 0.16.

For this we use the F test.

Numerator df= 5; denominator df= 988

H0: beta6=beta7=beta8=beta9=beta10=0

HA: H0 is not true.

F= [(0.4867 – 0.2541)/(1-0.4867)]*[988/5]= 89.54

The F is so large that at any conventional significance level, F>c, and the null will
be rejected. That is, the education dummies are statistically significant.

[5+15]

5. (a) To the explanatory variables in the regression in Qn 4(a), add land owned
(LandO) and land possessed (LandP) and re-run the regression. DO NOT paste
the output.

(b) Is either of the land variables individually significant at the 5 or 10% level?
Tland0= 0.000097/0.000077= 1.26
At 5%, c= 1.645. T<c, therefore do not reject null. Therefore, statistically
insignificant.
At 10%, c= 1.282. t<c, therefore insignificant.

TlandP= -0.000053/0.000077= -0.69


At 5%, t>-c, therefore fail to reject null. Therefore significant.
At 10% level, significant.

(c) Now drop land owned (LandO) and re-run the regression. Is the included
land variable significant at the 5 or 10% level?

We find the F stat. Unrestricted model is with the included land0 and restricted
is one without land0.

F= [(0.4802 – 0.4915)/(1 – 0.4802)]*[898/1]=

(c) Explain the pattern of results observed in (a) and (b).

[4+4+7]

Você também pode gostar