Question

Imagine that you regressed the earnings of individuals on a constant, a binary variable (“Male”) which takes on the value of 1 for males and is 0 otherwise, and another binary variable (“Female”) which takes on the value of 1 for female and is 0 otherwise. Because females typically earn less than males, you would expect:

Group of answer choices

autocorrelation or serial correlation to be a serious problem.

the estimated coefficient for Male to have a positive sign, and for Female to have a negative sign.

none of the OLS estimators to exist because there is perfect multicollinearity.

both coefficients to have the same distance from the constant, one above and the other below.

In a simple regression model, the value of the t-statistic, which tests the null hypothesis that the slope true slope coefficient is zero, is equal to 2. The F-statistic that tests the null hypothesis that the true coefficient of determination is zero is equal to:

Group of answer choices 4 0.5 2 6

Suppose that you estimate a multiple regression model using OLS using a sample of 120 observations. The skewness of the residuals is 0.5 and the excess kurtosis is 1. As a result, the value of the Jarque-Bera test is _______ and we ______ reject the null hypothesis of normally distributed disturbances at the 5-percent level of significance.

Group of answer choices 5; cannot 10; cannot 10; can 5; cannot

Answer 1

Ques 1:

none of the OLS estimators to exist because there is perfect multicollinearity.

Ques 2:

t² = F

2² = 4 = F

Therefore, F= 4

Ques 3:

JB = (n / 6) x[s2 + (k2 / 4)] Where, JB = Jarque-Bera Test Statistic n = Size of Sample s = Skewness Fact k = Kurtosis Fact

Therefore, we have

N = 120 , S = 0.5 and K = 1

Test statistic =

$\frac{120}{6} * ( 0.5^{2} + \frac{1^{2}}{4})$      =       10

Answer is 10; cannot