Econometrics

Question 1

MLR1-4

Answer 1

MLR.1 Linearity in parameters
MLR.2 Random sampling
MLR.3 Zero conditional mean E(u I x)=0
MLR.4 No perfect collinearity- any x cant be written as a linear comb of others

Question 2

MLR.5

Answer 2

Homoscedasticity: Var(u l x… )=sigma square (constant)

Question 3

Gauss-Markov

Answer 3

If MLR.1-5 holds then OLS estimators are BLUE

Question 4

If MLR.3 fails?

Answer 4

Unbiasedness fails

Question 5

If MLR.5 fails?

Answer 5

Efficiency fails

Question 6

If a relevant variable is omitted?

Answer 6

MLR.3 fails: Unbiasedness fails

Question 7

MLR.6

Answer 7

independent from explanatory variables and normality of u

Question 8

MLR.1-6

Answer 8

Classical assumptions: OLS estimators are overall the best (not just linear)

Question 9

SLR1-4

Answer 9

1. Linear in parameters
2. Random sampling
3. Zero cond. mean: E(u | x)=0
4. Sample variation in x: Var(x) not =0

Follows: unbiasedness!

Question 10

SLR 5

Answer 10

Homoscedasticity i.e.: Var (u|x)=sigma^2

Question 11

MLR.3 hangi durumlarda özellikle patlar?

Answer 11

1. Wrong fn form
2. Omitting an imp. var.
3. Measurement error in exp var.