3.5 mvsum
mvsum will return a consistently formatted table for different types of mutivariable models. This function now supports interaction terms
glm_fit = glm(Status~Sex:age+wt.loss,data=lung,family = 'binomial')
rm_mvsum(glm_fit)| Covariate | OR(95%CI) | p-value | Global p-value |
|---|---|---|---|
| wt loss | 1.00 (0.98,1.02) | 0.95 | |
| Sex:age | |||
| Female:age | 1.03 (0.99,1.06) | 0.14 | |
| Male:age | 1.05 (1.01,1.08) | 0.012 |
New Functionality & Bug Fixes
showNis an option to show the sample size for each variable/factor levelCIwidthallows for differ confidence intervals to be produced- automatically retrieve data from the model if not specified
- fixed bug with variable names embedded in level names
- fixed bug with centering in models
- added support for polr ordinal regression models
- updated CIs to use t-test instead of Z-test for means
- fixed treatment of glm objects to properly handle linear, binomial and poisson models
- fix to allow interaction terms in the model
TO DO
-allow for interactions with gee models