Multiple Regression Flashcards

1
Q

how to do stepwise regression

A
  1. predictor that has the highest correlation with outcome gets entered first
  2. next strongest correlation gets entered next
  3. stop when remaining predictos no longer improve prediction of the outcome
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

why stepwise is a bad idea

A
  • order that predictors are entered can depend on slight differences in predictive ability
  • slight numerical differences can lead to major theoretical differences
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

adjusted R2

A

gives us some idea of how well the model generalizes

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

ANOVA test what

A

whether the model is significantly better at predicting the outcoming using the mean as a best guess

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

F-ratio

A

ratio of improvmenet in prediction that results from fitting the model relative to the inaccuracy that still exists in model

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

how are standardized b-values labels

A

beta1

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

what does standardized b value tell us

A

the importance of each predictor (the bigger absolute value = more important)

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

how to assess multicollinearity

A

VIF

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

standarized residula

A

residuals converted to z-scores

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

standard residular that is an outliner

A

> 3,29

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

standard residular that’s unacceptable error

A

> 2,58 (1% of cases)

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

standard residual that indicates poor model

A

> 1.96 (5% of cases)

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

influential cases

A

exert undue influence on the parameters of amodel

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

how do you know if a case has undue influence

A

Cook’s D > 1

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

Leverage

A

influnece of a predictor value will be stronger when it is further from the mena of that predictor

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

mahalanobi’s distance

A

distance of cases from the mean(s) of the prediction

17
Q

covariance ratio

A

influence of a case on the variance of regression parameters

18
Q

assumptions in multiple regression

A

additivity and linearity
predictors uncorrelated with external variables
outcome must be continous
non-zero variance

19
Q

P-P plot

A

cumlative probability of standarized reisdual against cumulative probability of normal distribution