I have constructed a multivariate logistic regression model using lrm from the
Design library:
fit <- lrm(y ~ x1 + x2, data=mydata)
When I plot the summary of this fit, I get a very nice plot with shaded
confidence bars visually displaying the odds ratios of each covariate in the
model.
plot(summary(fit, vnames ='labels'))
The problem is that the resulting plot has a few things that I would like to
modify but am having a hard time doing:
1) Using the vnames option I can get the plot to put up the variables labels
that I like. However, when one of the covariates is a factor with multiple
levels, the plot shows names for the contrast levels that I do not like (ex: x2
- 1:0, x2 - 2:0). How can I change these to more meaningful names?
2) At the bottom of the plot the values of the different shades of the
confidence bars are plotted in microprint. How do I remove these values?
3) I would like to change the order of the variables on this plot, so they move
from the lowest OR to the highest. How do I do this?
I would very much appreciate any help with this issue.
__________________________________________________________
Dr. Brant A. Inman, MD FRCSC
Urologic Oncology Fellow, Mayo Clinic College of Medicine
200 First Street SW
Rochester, MN 55905
Phone: (507) 284-1330
Fax: (507) 284-4951
Email: inman.brant@mayo.edu
__________________________________________________________
|