Week 6

Question 1

if model SS> residual/error SS

Answer 1

good model if p also <.001

Question 2

hierarchical compression

Answer 2

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

Question 3

nested models

Answer 3

one includes all the variables of another model

Question 4

model specification

Answer 4

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

Question 5

log likelihood

Answer 5

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

Question 6

LL in linear models

Answer 6

negative LL is often SS deviations

Question 7

LL provides a direct measures

Answer 7

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

Question 8

AIC and BIC

Answer 8

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

Question 9

best model=

Answer 9

highest likelihood and fewest number of parameters/more parsimonious

Question 10

AIC

Answer 10

akaike information criterion
2k-2LL

Question 11

BIC

Answer 11

Bayesian information criterion
In(N)k-2LL