1
Q
What is the Wilks log likelihood ratio statistic
A
2
Q
What are the advantages of using a likelihood-based CI
A
- Not restricted to be symmetric
- Enables to construct multivariate CIs for parameter vector easily even in non-normal cases
- Contains normal CI as special case
3
Q
What is the quadratic approximation of the Wilks statistic
A
4
Q
Under π½0 how is the Wilks statistic distributed
A
5
Q
What are the critical values of the likelihood CI
A
6
Q
What is the likelihood ratio
A
7
Q
What is the relationship between the Wilk statistic and the likelihood ratio
A
π(π½0) = β2 log Ξ(π½0)
8
Q
How can the Wilks statistic be related to the KL divergence for specific models
A
Probability and Statistics 2
flashcards
Decks in class (18)
# Cards
-
1) Random Variables30
-
2) Multivariate Random Variables24
-
3) Expectation26
-
4) Generating Functions25
-
5) Sampling and Convergence15
-
6) Multivariate Normal Distribution15
-
1) Overview of Statistical Learning8
-
2) Entropy and Fisher Information19
-
3) Maximum Likelihood Estimation11
-
4) Quadratic approximation and normal asymptotics11
-
5) Likelihood-based confidence interval and likelihood ratio8
-
6) Optimality properties and conclusion10
-
7) Conditioning and Bayes rule8
-
8) Models with latent variables and missing data5
-
9) Essentials of Bayesian statistics3
-
10) Bayesian learning in practise14
-
11) Bayesian model comparison15
-
12) Choosing priors in Bayesian analysis1
