4- Hypothesis testing Flashcards

1
Q

What does ^βols tell us?

A

Marginal effect on the dependent variable for a unit change in the covariate

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

What do hats denote in regressions?

A

Estimated values, once we have estimated values of beta, the regression is now called the estimated regression equation

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

What is the estimated error (û) in terms of y?

A

û = y - ŷ

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

What is the explained sum of squares (ESS)?

A

Square of the difference between the estimated and the mean value Σ(ŷi - ȳ)²

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

What is the residual sum of squares (RSS)?

A

Square of the difference between the observed and estimated value Σ(yi - ŷi)²

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

What is the total sum of squares (TSS)?

A

Explained sum of squares + Residual sum of squares
Σ(yi - ȳ)²

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

What is R squared?

A

Variation of Y that is explained by the regression

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

What is R squared in terms of sum of squares?

A

ESS/TSS = 1 - (RSS/TSS)

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

What is the main flaw of R squared?

A

It increases as the number of covariates in the model increases, even if they are nonsensical variables

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

What is adjusted R squared?

A

It takes into account the number of parameters

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

What is the formula for adjusted R squared?

A

1 - (n-1)(1 - R²)/n-K

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

What is the null hypothesis?

A

Typically a statement that we do not expect (e.g. the coefficient is 0)

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

What is the alternative hypothesis?

A

Typically a statement that we expect (e.g. coefficient is not equal to 0)

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

What is the decision rule?

A

Method of deciding whether to reject the null hypothesis or not. This is done by comparing the sample statistic to the critical value

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

What is the critical value?

A

Threshold value. If the sample statistic is greater than (less than) the critical value we reject (do not reject) the null hypothesis

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

What is a Type 1 (false-positive) error?

A

When an investigator rejects a null hypothesis that is actually true in the population

17
Q

What is a Type 2 (false negative) error?

A

When an investigator fails to reject a null hypothesis that is actually false in the population

18
Q

What does the t-test check?

A

If the coefficients are significant i.e. non-zero

19
Q

What are degrees of freedom for a t test?

A

Observations minus parameters (n-K)

20
Q

What is the formula to run a t-test?

A

tₗ = ( ̂βₗ - H₀)/√( ̂vₗₗ)

21
Q

What is ̂vₗₗ in the t-test?

A

The relevant element from the variance matrix

22
Q

What are the 3 steps for a hypothesis test?

A
  1. Get the critical value from t table
  2. Plug given values into t test equation
  3. If t value is greater than critical value, reject null
23
Q

What are the 3 steps for the global significance test?

A
  1. Get critical value from fisher distribution
  2. Plug numbers into formula
  3. If result greater than critical, reject null
24
Q

What does the global significance test check?

A

Overall significance of model

25
Q

How can you find an upper and lower bound for an uncertain β based on a range of p values

A

-Figure out the lowest/highest p value the t stat must be greater/less than
-Form 2 inequalities using the t test formula
-Rearrange for an upper and lower value of β

26
Q

What is RSSᵣ & RSSᵤ?

A

RSS with restriction from null hypothesis & unrestricted original RSS

27
Q

What is the formula for an F-test?

A

[(RSSᵣ-RSSᵤ)/m] ÷ [RSSᵤ/(n-k)]

28
Q

What types of parameters can we not test using t or F tests?

A

Polynomial parameters; the tests only work on linear hypotheses

29
Q

When could you test regression coefficients using critical values from the normal distribution, not from the t distribution?

A

If n is extremely large, it might be appropriate to use the normal distribution, because the t(n-k) distribution converges on N(0,1) when n→∞

30
Q

What do you need to know to conclude whether a t-test is appropriate?

A

The distribution