Pictures Flashcards Preview

ECONOMETRICS > Pictures > Flashcards

Flashcards in Pictures Deck (50)
Loading flashcards...
1
Q

What is the estimate for αhat in this equation: yt=αyt-1+ut

A
2
Q

what is the weighted least squares model

A

E(ui^2|xi)=σ^2h(xi)

3
Q

Equation for F test based on R^2

A
4
Q

Equation for Logit

A
5
Q

Equation for probit model

A

φ(.) is the standard normal cumulative distribution function

6
Q

what is the densitiy of the logit

A
7
Q

what is the marginal effects of the Logit

A
8
Q

what is the marginal effects of the Probit

A
9
Q

what is the probability P(X=x) for a Bernoulli distribution

A
10
Q

what is the little tricks that help you derive the OLS estimator

A
11
Q

what are the conditions for Pooled OLS to be consistent and normally distributed in large samples

A
12
Q

what are the conditions for random effects GLS to be consistent and normally distributed in large samples

A
13
Q

what are the conditions for Fixed Effects OLS to be consistent and normally distributed in large samples

A
14
Q

what are the conditions for FIirst Differenced OLS to be consistent and normally distributed in large samples

A
15
Q

what is the random effects GLS conditional variances with strong exogeneity assumption E(ui|Xi)=0

A
16
Q

What is the estimator for the variance of βRE given conditional homoskedasticity and no serial correlation of uit, E(uituis|Xi)=0

A
17
Q

what is the between estimator

A
18
Q

What is the fixed effects OLS equation that takes into account endogeneity

A
19
Q

What is the further assumption on Fixed Effects OLS to make it efficient

A

That the uit are further conditionally homoskedastic and not serially correlated. uiui’ because its a matrix

20
Q
A
21
Q

what is the setting for the first differenced OLS

A
22
Q

What is the process of first differenced OLS

A
23
Q

Given this equation what is the condition for the OLS estimator for β to be consistent and normally distributed in large samples

A
24
Q

Given this equation, assuming there is feedback from ui,t-1 to xit, show that the FD OLS model is biased and inconsistent

A
25
Q

How is the endogeneity problem of the fixed effects OLS solved

A

First differenced model is a good starting point. Then use xi,t-j as an instrument as it satisfies the exogeneity condition as it is not correlated with (uit-ui,t-1) and is clearly correlated with the endogenous variable (xit-xi,t-1) as x responds to past realised shocks

26
Q

what is the density of the standard logistic distribution

A
27
Q

what is the density of the standard normal distribution function (probit)

A
28
Q

what are the marginal effects of the logit model

A
29
Q

what is the marginal effect of the probit model

A
30
Q

what is the maximum likelihood estimation for the probit model

A

Maximum likelihood for the probit: the conditional distribution yi | xi is Bernoulli

31
Q

what is βhat in matrix form

A
32
Q

how do you work out the conditional log-likelihood function (with picture)

A

write down the conditional density, then take logs

33
Q

what is the derivative of the logistic function

A
34
Q

what is the derivatives of logL wrt beta for logit

A

derivation ex lec 6

35
Q

what is the derivative of the logL wrt beta for probit

A
36
Q

what is the conditional log-likelihood function

A
37
Q

what is the density of a random variable

A

the derivative of its sumulative distribution function

38
Q

what is the derivative of LogL wrt beta for the probit model

A
39
Q

what is the AR(1) correlogram for equation yt=ρyt-1+et

A
40
Q

is AR(1) weakly dependent

A

AR(1) is weakly stationary and automatically satisfies weak dependence because Var=

41
Q

equation for an f test

A
42
Q

what is the White robust variance estimator

A
43
Q

equation for the average marginal effect of probit

A
44
Q

what is the marginal effects of probit at means

A
45
Q

average marginal effect of probit when binary variable

A
46
Q

what is the hausman test

A
47
Q

what is the covariance in an AR(1) process

A
48
Q

how do you get to the covariance for an AR(1) process

A

substitute for ytyt-1 then ytyt-2, but don’t substitute the whole equation in

49
Q

what is the variance covariance matrix of an AR(1) process

A

E(εtεt-j)=

50
Q

what is the variance covariance matrix for heteroskedasticity

A