L2 Flashcards

1
Q

What is Bias?

A

Difference between expected value of an estimator and the population parameter that it is estimating

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Difference between B1 and ^B1:

A

B1 is a constant

^B1 is a random variable with sampling distribution: E(^B1) and Var(^B1)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What makes an estimator unbiased?

A

E(^B) = B

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Draw the bias table

A

Corr
Bias

> <
<>

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What is the Kitchen Sink Regression problem?

A

Temptation to include all available variables.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Irrelevant Variables

A

Have no effect on unbiasdeness, can affect variance of OLS estimators

Most important when selecting variables is theorectical relevance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Consider the model:

wage = B0 + B1 female + u

What is B0?

A

Avg wage in the sample for men

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Consider the model:

wage = B0 + B1 female + u

What is B1?

A

Difference in avg wages in sample between men and women

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Consider the model:

wage = B0 + B1 female + u

What is B0 + B1?

A

Avg wage in sample for women

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Consider the model:

wage = 0.62 - 2.27 female + 0.51 educ

Why is B1 negative?

A

We can believe there are ommited variables, makes no sense if evertyhing else is 0 for a women to have a negative wage

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Collinearity in dummies

A

Cannot include dummies for all model groups (so choose base group)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Consider the model:

log(wage) = B0 + B1 married + B2 female + B3 (married x female)

What is B0?

A

Intercept for base group, in this case -> single male

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Consider the model:

log(wage) = B0 + B1 married + B2 female + B3 (married x female)

What is B0 + B1?

A

Intercept for married men

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Consider the model:

log(wage) = B0 + B1 married + B2 female + B3 (married x female)

What is B0 + B2?

A

Intercept for single women

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Consider the model:

log(wage) = B0 + B1 married + B2 female + B3 (married x female)

What is B0 + B1 + B2 + B3?

A

Intercept for married women

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Consider the model:

log(wage) = B0 + B1 married + B2 female + B3 (married x female)

What is B3?

A

Difference in marriage premium between men and women

17
Q

Larger u variance -> larger ^B1 variance

A

More variation in u -> harder to estimate B1

18
Q

Sum ( xi - avg(x) ) increases as N increases

A

Larger samples -> better estimations

19
Q

Sum ( xi - avg(x) ) reflects x variance

A

Larger x variance is good

20
Q

T-statistic tells us what?

A

How far the estimated parameter is from the true value

21
Q

Normality of the errors

A

Assume u ~ iid (0, sigma^2)

Large samples

22
Q

If H0: B1 - B2

Show t

A

t= Ô - O / se(Ô)

Ô = B1 - B2

se(B1-B2) != se(B1) - se(B2)

23
Q

What is the F-test for?

A

Understand how much worse does the restricted model fit the data, when compared to the unrestricted

24
Q

What does the F-test measure?

A

Change in SSR between models

25
F-test formula
F = SSR (R) - SSR (UR) / SSR (UR) cheat sheet
26
Is F always larger than 0?
Yes
27
F-stat for parameter exclusion is equal to square of t-stat in what case
1- parameter exclusion (2-sided H)
28
3 types of Functional Form Misspecification
1 - Omitting higher powers of variables 2 - Omit variable interactions 3 - Using levels instead of logs
29
What does FFM lead to?
Biased OLS coefficients ( can be tested )
30
What does RESET test consist of?
Estimate model Compute predicted values See if predicted values in different powers have significance
31
What does Davidson and Mckinnon test consist of?
Compare two models, e.g.: y = B0 + B1 x + u y = B0 + B1 log(x) + u Use fitted values of other in each: y = B0 + B1 x + B2 (^y) y = B0 + B1 log(x) + B2 (^y) Where the fitted are not significant in the other model, we reject
32
If after Davidson and Mckinnon we cannot decide between models, what is the tie-breaker?
R-squared