exam issues Flashcards

1
Q

What is the relationship between t test and f test?

A

f test is the square of a t test.

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

What is the coefficient of determination?

A

R^2

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

How do you estimate the slope coefficient using OLS

A

ybar - b1bar(xbar) must be using the means!

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

What is the OLS estimate of B0 if it is just a B0 in the model?

A

It is just ybar

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

How would I calculate R^2 from data?

A

You do TSS which is sum of (yi-ybar)
Then work out ESS (ybar hat - ybar)^2

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

What is the coefficient of determination?

A

R^2

yi-ybar and square all these

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

What is an alternative way to work out SE of regression?

A

sqrt RSS/DOF

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

How do you get R^2 from correlation coefficient?

A

r^2 = R^2

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

What does it mean that the expectation of the error term of 0

What does the constant variance mean?

A

On average the error terms are equaled to 0

-does not mean that small x has a different level of uncertainty compared to big x

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

If OLS assumptions are satisfied what do the estimates?

A

Estimates have smallest variance and are unbiased.

What assumption is needed.

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

When does TSS change?

A

When dependent variable changes

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

What is a unit change in a %?

A

A % point change

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

What is the t-ratio?

What does it tell us?

A

Coefficient divided by the standard error

The closer to 2, the more significant.

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

What is the standard error of the regression?

A

You can get it from MS which is the error of the error term.

Then take the square root.

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

What is the RMSE?

A

Root mean squared error
sqaure root of MS of residuals

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

What are the ways that d can be interpreted?

A

a) the equality signs in beta
b) the difference in parameter between restircred and unrestricted model

17
Q

When you are testing that all slope coefficients are equal to 0 what is the RSSR?

A

Just the TSS of the unrestricted model

18
Q

For a single restriction what is the relationship between f test and t test?

A

f test is square of t test

19
Q

What is the F test trying to say?

A

That the model explains some of the variatino

20
Q

What is a pvalue?

A

Proability of observing a test statistic given H0 is true

21
Q

Graphically what is the power of a test?

A

-The proportion of the given region which is in the rejection region.

22
Q
A