Econ Case

Name: Chun Yat Kwok        Student ID: 951389819

II-1.

[pic 1]

The maximum value of earnings is 197.23

The standard deviation of asvabc is 9.522961

II-2.

Histogram of the earnings:

[pic 2]

Histogram of the asvabc:

[pic 3]

Histogram of l_earnings:

[pic 4]

III-1.

Scatterplot of the explanatory variable against the residuals:

[pic 5]

Based on the scatterplot, it does not look homoskedastic but heteroskedastic because the variance of disturbance term is increasing as ASVABC increases.

III-2.

[pic 6]

Based on the scatter, it looks like homosckedastic because the variance of the disturbance term is almost the same for all observation.

III-3.

gen scaled_l_earnings = l_earnings / l_asvabc

gen inv_asvabc = 1 / l_asvabc

[pic 7]

Based on the scatter, it looks like homosckedastic because the variance of the disturbance term is almost the same for all observation.

IV-1.

In Stata:

regress earnings asvabc if _n<4456

[pic 8]

regress earnings asvabc if _n>(_N-4455)

[pic 9]

1. The null hypothesis is the beta_2 is 0
2. The test statistic, RSS2/RR1 is  6.489917636
3. The degree of freedom  of F is 1 and 4453
4. F critical value at 5% and 1% are 3.85 and 6.66
5. Therefore, we can reject the null hypothesis at 5% level because the test statistic is larger than the critical value at 5% level; however, we fail to reject the null hypothesis at 1% because it is smaller than the critical value at 1% level.

IV-2.

In Stata:

regress l_earnings l_asvabc if _n<4456

[pic 10]

regress l_earnings l_asvabc if _n>(_N-4455)

[pic 11]

1. The test statistic, RSS2/RSS1 is 1.522719047
2. We fail to reject the null hypothesis at 1% and 5% because the test statistic is smaller than the critical value at 1% and 5%.

   ---

IV-3.

In Stata,

regress scaled_l_earnings inv_asvabc if _n<4456

[pic 12]