Deviance for Binary Data Notes

P value for d p d dobs obtained from 2dof unm other

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: of covariate combinations − no. of estimated regression coefficients. In general, large values of D imply model lack of fit. D not testing for Binomial assumption of the data. D testing if one or more predictors have been omitted from the model. p-value for D : P [D > Dobs ] obtained from χ2dof ) . ( UNM Other ’diagnostics’ (summaries) pseudo R 2 statistic, pseudo R 2 = ˆ logLs − logL(β ) logLs where LS is the max. likelihood for the saturated model ˆ and L(β ) is the max likelihood for a model with covariates. ”proportional improvement in log-likelihood”. Another pseudo R 2 statistic (Mc Fadden’s) ˆ ˆ logL(β0 ) − logL(β ) pseudo R 2 = ˆ logL(β0 ) Efron’s R2 = 1 − N i =1 (yi N i =1 (yi − πi )2 ˆ ¯ − Y )2 UNM Residuals Yi number of successes. ni number of trials. πi estimated probability of success based on a glm ˆ Pearson chi-square residuals ri = Yi − ni πi ˆ ni πi (1 − πi ) ˆ ˆ ; i = 1, 2, . . . , N Chi-square statistic, N X2 = ri2 i =1 has the same dofs as D , N − (p + 1). How does deviance work for a Poisson regression? UNM...
View Full Document

Ask a homework question - tutors are online