Applied Econometrics - Heteroskedasticity in Statistics

Economics
Student
Applied Econometrics
Student

Economics
Student
Question 1
Section A
1. a) Never make any difference
2. d). Biased but the bias usually falls as sample size increases.
3. The Gauss Markov Theorem does not hold.
4. Converge but a similar rate to that of stationary process.
5. Auto-correlation
Question 2
b. TEST 1: Jarque-Bera Test
H0: Data In the variable is normally distributed
H1: Data in the variable is not normally distributed
Results: Chi-square: 3.95104, p-value: 0.13871
Decision: We fail to reject the null hypotheses and therefore conclude that the data is normally
distributed.
TEST 1: Breusch-Godfrey LM Test
H0: There is no serial correlation of any order up to p

Economics
Student
H1: There is a serial correlation of any order up to p
Results: F-statistics: 2.70145, p-value: 0.024549
Decision: Fail to reject the null hypothesis.
TEST 1: Ramsey Reset Test
H0: The coefficient is equal to zero
H1: The coefficient is not equal to zero
Results: F-statistics: 3.28950, p-value: 0.07293
Decision: Reject the null hypothesis.
TEST 1: ARCH Test
H0: No ARCH effect present
H1: ARCH effect present
Results: F-statistics: 1.26877, p-value: 0.28823
Decision: fail to reject the null hypothesis.
TEST 1: White Test
H0: The variances for the errors are equal