Logistic regression: improve existing documentation

[andreaczhang]

• understand whether the data example is appropriate
• move R part in the comparison page in the appropriate place
• if needed, re-run the examples

[andreaczhang]

• 2 new variables: 2 levels: tx cx; 3 level cx, dose A, dose B (among the active ones)
• lung_cancer.csv

analyses: two analyses

Update this section to be 1) 2 trt geting out odds ratio using confint.default and confint as described above. Then repeat using 2) 3 trt model & how to do contrasts of 2 trts vs a third treatment as per this code below.

Contrast statements for 2 or more treatments

To create contrasts, you can use the fit.contrast() function from the gmodels package.
This can be used with lm and glm objections:
Suppose we had a 3 level treatment variable (trt01p), whose levels were ordered Dose1, Dose2, Placebo.
You would fit the model as above, followed by fit.contrast(). This is effective testing the null hypothesis that 0.5dose1 + 0.5 dose2 - placebo = 0.
m2 <- glm(wt_grp ~ sex + trt01p, data = lung2, family = binomial(link=“logit”))
fit.contrast(m2,‘trt01p’,c(0.5,0.5,-1),conf.int=0.95)