Dear S list members, I have a dataframe called mydata with column names c("y1", ...,"y400","group","male","age","Education.Level") I wrote a loop to run a series of regressions with response variable
I want to thank the helpful suggestions from Max Zhao, Bill Dunlap, Dimitris Rizopoulos and Shawn Boles. Max suggested to use option "singular.ok=T" to deal with collinearity. Bill suggested to use x
I guess that R^2 from the intercept model is always smaller than the R^2 from the non-intercept model. Below is what I thought. Please correct me if I misunderstood anything.. Let SSR = regression s
The event is actually "the present of a symptom". Patient's visits are not equally spaced. We are looking at the first time the patient has the symptom, and how long it takes the symptom will disapp
Many thanks to Dimitris Rizopoulos and Sam Buttrey who suggested the use of split and lapply (or sapply). My original question is: I have a very large cohort data that contains repeated measures. The
Dear All, When I fit a logistic regression, say, using out<- glm(y ~ x1 + x2, family = binomial(link = logit)) How can I find what type of residuals that out$residuals gives me? How can I check (what
Many thanks to professor Brian Ripley, Sundar Dorai-Raj and Peter Alspach for their quick responses to my question. The suggestions are very helpful. Thanks again! Regards, Peng Huang Dear All, When
Dear All, I have a very large cohort data that contains repeated measures. The data looks like: id days event 1 0 0 1 30 0 1 50 0 1 95 1 1 120 1 1 147 0 1 180 0 2 0 0 2 75 0 2 100 0 3 0 0 3 37 1 3 80
Dear All, Could some one point out to me some references on rank based models or tests for multivariate outcomes measured longitudinally? All outcomes are correlated and repeated measures are depende
x _ 2*pval$p.value -- Peng Huang, Ph.D. Assistant Professor Department of Biometry and Epidemiology Medical University of South Carolina 135 Cannon Street, Suite 303 Post Office Box 250835 Charlesto
If there are more than two observations per unique id, we may try the following: id x 1 2 33 2 2 NA 3 3 67 4 3 NA 5 3 NA 6 3 NA 7 3 NA 8 4 24 9 4 NA 10 4 NA 11 4 NA 12 5 45 13 5 NA 14 5 NA 15 5 NA 16
I guess the reason you got error message from my code is because the code can not be run for the second time. Here is why. Let's create a data set: id <- sample(1:3,10,T) id<- sort(id) x1<- rle(id)