Download as pdf or txt
Download as pdf or txt
You are on page 1of 7

Statistical Properties of OLS

Income (x) and Consumption (y)

Gujarati (2004), Basic Econometrics


Income (x) and Consumption (y)

Gujarati (2004), Basic Econometrics

Assumptions required for SLR

• SLR A1: Linear in the parameters

• SLR A2: Random sample {(xi,yi): i=1, …,n}

• SLR A3: E(u|x) = 0

• SLR A4: There is some variation in x in the


population.
Unbiasedness of OLS
n

∑ (x − x )( y
i i − y)
βˆ1 = i =1
n

∑ (x − x )
2
i
i =1

(xi − x ) yi
βˆ1 = ∑ , where s x2 ≡ ∑ ( xi − x )
2

s x2
Variances of the OLS estimator

• SLR A5: Homoskedasticity - Var(u|x) = σ2

Homoskedastic Case
y
f(y|x)

. E(y|x) = β + β x
0 1
.

x1 x2
Heteroskedastic Case
f(y|x)

y
.
. E(y|x) = β0 + β1x

.
x1 x2 x3 x
Sampling Variance of OLS estimator

( ) 1
Var βˆ1 = Var ( β1 + ( 2
sx
∑ (xi − x )ui ), where s x2 ≡ ∑ ( xi − x )
2

Estimating the Error Variance


• We don’t know what the error variance (σ2 ).
• We observe only the residuals, ûi

uˆi = yi − βˆ0 − βˆ1 xi


= (β 0 + β1 xi + ui ) − βˆ0 − βˆ1 xi
i (
= u − βˆ − β − βˆ − β
0 0 ) ( 1 1 )
Unbiased estimator of σ 2 :

i = SSR / (n − 2 )
1
σˆ =
2

(n − 2) ∑ ˆ
u 2

You might also like