11) Bayesian model comparison Flashcards
What is the marginal likelihood in Bayesian learning of a model class M
What distinguishes a “simple” model from a “composite” model
How is the likelihood of a composite model M computed
How can the log-marginal likelihood be expressed
How does Bayesian inference account for model complexity
Penalty for Complexity: Log-marginal likelihood includes a penalty for the number of parameters
Occam’s Razor: The principle of preferring a less complex model
What is the Bayes factor
The ratio of the likelihoods of the two models
How do you compare two models M1and M2 using Bayes Factor
What is the orginal scale for Baye’s Factor
What is the modified orginal scale for Baye’s Factor
What is the Bayes factor for simple models
If both M1 and M2 are simple models, the Bayes factor is identical to the likelihood ratio of the two models
What is the Schwarz approximation of the log-marginal likelihood
What is the Bayesian Information Criterion (BIC) for a model M
How can the log-Bayes factor be approximated using BIC
What are true and false positives and negatives in the context of statistical testing
True Positive (TP): Elements correctly identified as positive.
True Negative (TN): Elements correctly identified as negative
False Positive (FP): Elements incorrectly identified as positive
False Negative (FN): Elements incorrectly identified as negative
What are the definitions of True Negative Rate (TNR), True Positive Rate (TPR), and Accuracy
