Lecture 2

Question 1

Linear regression with time series data

Answer 1

model assumptions
method of Ordinary Least Sqaures
Method of momednts
Properties of OLS in finite and in infinte samples

Question 2

Model assumptions

Answer 2

Can be regressed as
* Yt = Bo +B1 xt + et
if we assume E [et|xt] = 0, then
* E [yt|xt] = Bo +B1 xt + E [et|xt]
* Bo +B1 xt

happens as the condtional mean u= 0, the unconditonal u will be 0
happens as et is true random

Question 3

Homoscedasticty

Answer 3

Variance et is constant for all values of x

Question 4

asummptions beteween xt and et

Answer 4

crucial for interpreting the model and derving properties of any estimator Bo and B1

Question 5

Overview of model assumptions- Normal regression w/fixed regressor

Answer 5

E [y|x] = XB
X = fixed (not random)
Rank (x) = full
e= fully independent , homoscedastic, no serial correlation I.e. σ^2I , et~ N(0,σ^2)
B^ols ~ N, unbiased (BUE)

Question 6

Overview of model assumptions- Normal regression w/stirctly exog. regress

Answer 6

E [y|x] = XB
X = stochastic
Rank (x) = full
e= mean independence with E[ et|xt] = 0 ∀t, homoscedastic , np serial corelation = Var (e|x) =σ^2I , et|x~ N(0,σ^2)
Bols|x~N , unbiased

Question 7

Overview of assumption models- regression w/weakly exog. regress

Answer 7

E [y|x] = XB
X = stochastic
Rank (x) = full
e= mean independence with E[ et|xt] = 0 ∀t, homoscedastic , np serial corelation = Var (e|x) =σ^2I , et|x~ N(0,σ^2)