1 Matching Annotations
  1. Feb 2017
    1. This is sometimes called a likelihood ratio test (the deviance residual is -2*log likelihood)

      see logLik() in R, and note that all return the same value, which is chi-square distributed, and can be ways to select the appropriate model. Ways to compare models and variable selection:

      1. (logLik(fit.1) - logLik(fit.2)) * -2
      2. anova(fit.1, fit.2)
      3. fit.1$deviance - fit.2$deviance