Week 6 Flashcards

1
Q

if model SS> residual/error SS

A

good model if p also <.001

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

hierarchical compression

A

special care of model comparison when the two models being compared are nested

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

nested models

A

one includes all the variables of another model

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

model specification

A

picking what goes into the model including form (linear vs non linear) and what parameters to icnlude

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

log likelihood

A

most models work by maximising the LL
trying to find a peak that uses parameters that will maximise the likelihood of getting observed samples
Higher LL= better goodness of fit of that model

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

LL in linear models

A

negative LL is often SS deviations

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

LL provides a direct measures

A

how likely is it that we could observe these data given some parameter estimate?

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

AIC and BIC

A

measures of a good model that includes a penalty for model complexity
help selevt best model
lower scores are better, pick the model (M1, M2) with lowest AIC/BIC

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

best model=

A

highest likelihood and fewest number of parameters/more parsimonious

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

AIC

A

akaike information criterion
2k-2LL

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

BIC

A

Bayesian information criterion
In(N)k-2LL

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