SU5 - Further Issues In Linear Regression: Modelling And Inference Flashcards

1
Q

What is R squared?

A

Statistic for evaluating if the model fits the data well

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

4 Properties of R squared?

A

1) R squared cannot be negative.
2) R squared is bounded between 0 and 1
3) R squared = 0 if SSE = 0
4) The 𝑅2 is non decreasing as you add more explanatory variables into the model

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

Why R squared cannot be negative?

A

because SSE and SST, sums of squared terms cannot be negative

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

Why R squared = 1 if SSR = 0?

A

SST = SSE

Variation in Y is completed accounted by variation in Y(hat). In this case, Y(hat) fits the data perfectly.

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

Why R squared = 0 if SSE = 0?

A

R squared = 0 if Y(hat) has no variations. If Y(hat) has no variations, it does not explain Y at all.

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

Does R square decrease?

A

Never decreases, usually increases when another independent variable is added to a regression. Thus it is a poor tool for deciding whether one variable or several variables should be added to a model

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

What is a regression through the origin?

A

A regression without an intercept term

ln(1+π‘‘π‘Žπ‘₯)=πœƒ1ln(1+π‘–π‘›π‘π‘œπ‘šπ‘’)+𝑒 where β€œ1” is added to tax and income to prevent the log of zero.

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

Why is regression through the origin a bad idea?

A

1) If intercept is really zero, no harm adding in the intercept. If intercept is not zero, then estimates of both intercept and slope coefficient will be wrong
2) With regression through the origin, it is possible for R squared to be negative even though R squared should be between 0 and 1

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

What happens if we include an irrelevant variable?

A

The unbiasedness of the regression will not be affected but the variance could be affected

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

What happens if we omit a relevant variable?

A

Generally causes the OLS estimators to be biased (omitted variable bias) or worse, inconsistent

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

How to deal with multicollinearity?

A

1) Increase sample size

2) drop some variables that are highly collinear

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

What does multicollinearity affect?

A

Will not affect the variances of all OLS slope estimators. Only those from highly correlated regressors are affected

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

What happens if Xj is highly correlated with one or more regressors in the model?

A

R2 will be very high and Var will be large, causing πœƒπ‘— to be imprecise

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

What is the error normality assumption?

A

The population error e is independent of the explanatory variables and is normally distributed with zero mean and variance.

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

What is the linear model called when it is under six assumptions?

A

The classical linear model (CLM)

Assumptions of linear regression 1-5 + error normality

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

What is the population model expressed as when under CLM assumptions?

A

π‘Œ|π‘‹βˆΌπ‘π‘œπ‘Ÿπ‘šπ‘Žπ‘™(πœƒ0+πœƒ1𝑋1+πœƒ2𝑋2+β‹―+πœƒπ‘˜π‘‹π‘˜,𝜎2)

OLS have an exact normal distribution

17
Q

In Var(Λ†ΞΈj)=Οƒ2/SSTj(1βˆ’R2j), Why is Οƒ2 unknown?

A

o2 is the variance of e and we do not observe what e is.

18
Q

Why do we use n-k-1 rather than n in the denominator.?

A

estimated o2 will be an unbiased estimator of o2

19
Q

What is an unrestricted model?

A

It’s a model containing all the regressors prior to making the hypothesis

20
Q

what is a restricted model?

A

it is the model if H0 were true

21
Q

How to determine if H0 is true for multiple hypothesis testing?

A

Comparing the error term of unrestricted model vs the error term of the restricted model

22
Q

If H0 is true, what would the SSR of both be like?

A

SSR (restricted) would be the same as SSR (unrestricted) since the residuals should be the same

F statistic would be small

23
Q

If H0 is untrue, what would the SSR of both be like?

A

SSR (restricted) would be much larger than SSR (unrestricted)