Lecture 6: Logistic Regression (Alternative) Flashcards

(35 cards)

1
Q

What type of outcome does linear regression model?

A

Continuous outcome

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

What type of outcome does logistic regression model?

A

Categorical outcome

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

What assumption does linear regression make about relationships?

A

Linear relationship

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

Can logistic regression handle non-linear relationships?

A

Yes

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

Does logistic regression require normally distributed predictors?

A

No

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

What function does logistic regression use to model probabilities?

A

Logistic regression uses the logistic (or sigmoid) function to model probabilities.

This transforms the linear combination of predictors into a value between 0 and 1, which can be interpreted as a probability.

  • It ensures valid probability predictions.
  • It models the log odds of the outcome as a linear function of the predictors.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is the range of values produced by the logistic function?

A

0 to 1

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

What must outcome categories be in logistic regression?

A

Mutually exclusive and exhaustive

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

What happens if outcomes are not mutually exclusive or exhaustive?

A

Model assumptions are violated

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

What are logistic regression coefficients initially expressed as?

A

Log odds

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

What are odds in logistic regression?

A

Ratio of event likelihood to non-event likelihood

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

What does exponentiating a coefficient b yield?

A

Odds ratio (e.g., if exp(B) = 1.5 → each extra date increases odds of being in a relationship by 50%)

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

What does exp(b) = 1.5 mean?

A

Odds increase by 50% per unit increase in predictor

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

Are changes in probability constant in logistic regression?

A

No

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

Where is the probability change largest in a logistic curve?

A

Near 0.5

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

What does b = 0 mean in log odds interpretation?

A

No effect of predictor

17
Q

What does exp(b) = 1 mean in odds ratio interpretation?

A

No effect of predictor

18
Q

What is the null hypothesis in logistic regression?

A

Predictor has no effect on the outcome

19
Q

What test checks if a coefficient significantly differs from 0?

20
Q

What type of distribution is used in the Wald Test?

A

Chi-square distribution

21
Q

What does it mean if the CI for exp(b) excludes 1?

A

Predictor is significant

22
Q

What does it mean if the CI for exp(b) includes 1?

A

Predictor is not significant

23
Q

What is a pseudo-R² used for in logistic regression?

A

Assessing model fit

24
Q

Name two types of pseudo-R².

A

Cox & Snell, Nagelkerke

25
What test evaluates model classification accuracy?
Hosmer-Lemeshow test
26
What does a significant Hosmer-Lemeshow test indicate?
Poor model fit
27
What does the omnibus test evaluate?
Whether all predictors improve the model
28
What method adds all predictors at once?
Direct (Enter) method
29
What method adds predictors in blocks?
Sequential (Hierarchical) method
30
What method selects predictors based on statistical criteria?
Stepwise method
31
What does forward stepwise start with?
No predictors
32
What does backward stepwise start with?
All predictors
33
What does each step in stepwise regression involve?
Model comparison
34
What type of outcome does multinomial logistic regression model?
Outcomes with more than two categories
35
How does multinomial logistic regression handle multiple categories?
By using dummy coding