lara harrup (IAH-P)
2009-Jun-24 17:55 UTC
[R] Goodness of fit test / pseudo r^2 measure for Zero Inflated Model
Hi I have been using a Zero-Inflated negative binomial model fitted using the pscl zeroinfl command but I would like to extract a goodness of fit measure are there any suitable pseudo R^2 measures available for this type of analysis to try and assess the amount of variation in the data explained by the model? I have tried with the pR2 command in pscl (for computing various pseudo R2 measures), the same package as zeroinfl, but I get the message that there is no-applicable method when I try it with a zeroinfl model. Any suggestions most appreciated. Many thanks in advance Lara [[alternative HTML version deleted]]
Apparently Analagous Threads
- Multiple Comparisons for (multicomp - glht) for glm negative binomial (glm.nb)
- Cross-Validation for Zero-Inflated Models
- randomForest - what is a 'good' pseudo r-squared?
- Prediction intervals for zero inflated Poisson regression
- Zero-inflated neg.bin. model and pscl package