Financial Econometrics. In-Class exercise.
Name (ID):
March 1, 2024
1. Simple linear regression assumptions:
SR1 : yi = β1 + β2 xi + ϵi , i = 1, . . . , N
SR2 : E(y|x) = β1 + β2 x
SR3 : var(y|x) = σ 2
SR4 : cov(yi , yj ) = 0
SR5 : The variable x is not random and must take at least two different values.
SR6 : ei |x ∼ N (0, σ 2 )
2. Explain what condition the least square estimates need to satisfy?
Solution:
Define the sum of squares: X
S(β1 , β2 ) = (yi − βi − β2 xi )2
Least squares estimate b1 , b2 satisfy
b1 , b2 = arg min S(β1 , β2 )
β1 ,β2
3. Write down the formula of the least square estimates.
Solution: P
(xi − x̄(yi − ȳ))
b2 = P ; b1 = ȳ − b2 x̄
(xi − x̄)2
4. Write down the sampling distribution of the least squares estimators.
Solution: If SR6 holdes, we have
σ 2 x2i
P
b1 ∼ N (β1 , P )
(xi − x̄)2
σ2
b2 ∼ N (β2 , P )
(xi − x̄)2