Section 4 Flashcards Preview

Econometric methods > Section 4 > Flashcards

Flashcards in Section 4 Deck (23)
Loading flashcards...
1
Q

What does section 4 consider?

A

Violation of var(ε)=σ^2I(n)

ie. violating assumption of errors being homoskedastic and not autocorrelated

2
Q

What does var(ε)=?

and so how is this when not violating A2?

A

E(εε’)= nxn matrix with variances down main diagonal and autocovariances in all other elements (see notes)

Under homoskedasticity, all of the main diagonal elements are the same σ^2, and all the autocovariances are equal to 0 tf var(ε)=σ^2I(n)

3
Q

What does var(ε) equal when violating A2?

A

var(ε)=σ^2Ω, where Ω is not equal to I(n)

4
Q

What are non-spherical disturbances? (2)

A

If in the var(ε) matrix, either:
a) the main diagonal elements aren’t constant (heteroskedasticity)
or
b) the off diagonal elements aren’t all 0s (autocorrelation)

5
Q

Is the OLS estimator still unbiased when A2 is violated?

A

Yes - see S2.3 for proof (doesn’t use A2 in it)

6
Q

Is the OLS estimator still efficient when A2 is violated? And note?

A

No - see S2.3 derived variance matrix of b, but used A2 to derive it
Furthermore, when proving variance of estimator was smallest amongst all unbiased estimators used A2 aswell

7
Q

Derive the heteroskedastic and AC consistent variance estimator?

A

Same as before up til:

var(b)=(X’X)^-1X’E(εε’)X(X’X)^-1 then just simplify (see notes for proof)

8
Q

What is a more efficient estimator than the heteroskedastic and AC consistent variance estimator?

A

GLS estimator

9
Q

Explain how a GLS estimator works?

A

Works by transforming the regression model into one with no heteroskedastic or autocorrelated errors.
The new model is then estimated by OLS like normal

10
Q

Given var(ε)=σ^2Ω, what do we know about Ω? (4)

A

nxn (square), +ve definite, invertible and symmetric

11
Q

Given var(ε)=σ^2Ω and its properties, what does this therefore mean? How does this help us develop GLS?

A

Means there is an nxn matrix P such that:
Ω^-1=P’P
This is called a Cholesky Decomposition and P is essentially equivalent to the square root of the matrix Ω^-1
By multiplying the regression model by P, we can eliminate the HTK and AC issues

12
Q

ε*=?

A

Pε - doesn’t violate A2!

13
Q

Prove ε* doesn’t violate A2?

A

See notes page 1 side 2

14
Q

Prove the formula for b(GLS)?

A

See notes page 1 side 2

15
Q

Show that b(GLS) is unbiased?

A

See notes page 1 side 2

16
Q

Derive var(B(GLS)) from equation before?

A

See notes page 1 side 2

17
Q

Derive var(B(GLS)) from first principles?

A

See S4 page 7

18
Q

Prove that the variance of the GLS estimator is smaller than that of the OLS estimator (use updated equation not original!)? (ie. prove GLS is more efficient)

A

See notes for putting it into form D=σ^2AΩA’ where:
A = (X’X)^-1X’-(X’Ω^-1X)^-1X’Ω ^-1

tf since Ω is a variance matrix tf is positive definite by definition. By pre and post multiplying it by A and A’, it is essentially the same as squaring it, and since σ^2 must be positive too since it is variance, by multiplying all these parts together can see that it must be positive definite aswell tf GLS is more efficient!

19
Q

See

A

page 8 summary of A2 being violated

20
Q

How to apply GLS for heteroskedastic errors?

A

See notes page 2 side 1 for WHY
The GLS transform is to divide the model by the square root of the variable in the error variance formula (ie. the root of the variable the error is a function of)

ie. doing PY since P = diagonal matrix with 1/sq.root of variable

SEE NOTES

21
Q

How to apply GLS for autocorrelated errors?

A

see notes aswell page 10

22
Q

What is the Prais-Winsten formula used for?

A

When doing GLS for autocorrelated errors, PY column represents the quasi-differences of the variables; apart from the first observation. The first observation is different because when calculating the quasi-difference of the first observation you need to know Y0 which you do not have (Y1 is first obs.) tf econometricians use the PW formula that avoids the loss of the first quasi-differenced observation

23
Q

Why do we need a different equation sometimes for feasible least squares (FGLS)?

A

Since often we don’t know Ω, for example, in the autocorrelation case we do not know the coefficient of autocorrelation ρ. Tf there needs to be an intermediate step in the estimation procedure that finds Ω(hat) and tf finds ρ(hat), such that the FGLS estimator becomes (see equations)