Multiple Regression Flashcards

1
Q

Logic:

A

Is there reason to expect a causal relationship between the predictors and the response variable.
• Like Correlation, Regression can’t prove causation – that assumption is on us as the data collectors/owners.

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

Fit:

A

Does the overall regression show a significant relationship between the predictors and the response variable?

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

Parsimony:

A

Does each predictor contribute significantly to the explanation? Are some predictors not worth the trouble?

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

Stability:

A

Are the predictors related to one another so strongly that regression estimates become erratic?

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

Occam’s Razor

A

When two explanations are otherwise equivalent, we prefer the simpler one

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

R2 Adjusted

A
  • The closer R2 adjusted is to R2 the more each predictor is contributing
  • The further apart the weaker the contributors
  • R2 Adjusted will always be less than R2
How well did you know this?
1
Not at all
2
3
4
5
Perfectly