GLM.jl
GLM.jl copied to clipboard
Add additional methods
- [ ] add1.lm*
- [ ] alias.lm*
- [x] anova.lm* ->
ftest(#182) - [ ] case.names.lm*
- [x] confint.lm*
- [ ] cooks.distance.lm*
- [x] deviance.lm*
- [ ] dfbeta.lm*
- [ ] dfbetas.lm*
- [ ] drop1.lm*
- [ ] dummy.coef.lm*
- [ ] effects.lm*
- [x] extractAIC.lm* ->
aic,bic - [x] family.lm*
- [ ] formula.lm*
- [ ] hatvalues.lm*
- [ ] influence.lm*
- [ ] kappa.lm*
- [ ] labels.lm*
- [x] logLik.lm* ->
loglikelihood - [x] model.frame.lm*
- [x] model.matrix.lm*
- [x] nobs.lm*
- [ ] plot.lm*
- [ ] predict.lm*
- [ ] print.lm*
- [ ] proj.lm*
- [ ] qr.lm*
- [x] residuals.lm*
- [ ] rstandard.lm*
- [ ] rstudent.lm*
- [ ] simulate.lm*
- [ ] summary.lm*
- [ ] variable.names.lm*
- [x] vcov.lm*
great use of the new task lists :-)
I needed an excuse to try them out.
See https://github.com/JuliaStats/StatsBase.jl/issues/280 about improving the consistency of our modeling API.
See JuliaStats/StatsBase.jl PR#355. It includes a few such as leverage. For rstandard and rstudent, those could be added to residuals with keyword arguments. We could add also add formula(::StatisticalModel).
Any way leverage and cooks distance will be added? (btw, anybody has an implementation that can share?)