|
|
|
||||
|
|
|||||
The R package(s) needed for this chapter is the survival package. We currently use R 2.0.1 patched version. You may want to make sure that packages on your local machine are up to date. You can perform updating in R using update.packages() function.
Table 1.1 on page 4, data set is hmohiv.csv.
hmohiv<-read.table("http://www.ats.ucla.edu/stat/R/examples/asa/hmohiv.csv", sep=",", header = TRUE)
attach(hmohiv)
hmohiv
ID time age drug censor entdate enddate 1 1 5 46 0 1 5/15/1990 10/14/1990 2 2 6 35 1 0 9/19/1989 3/20/1990 3 3 8 30 1 1 4/21/1991 12/20/1991 4 4 3 30 1 1 1/3/1991 4/4/1991 5 5 22 36 0 1 9/18/1989 7/19/1991 6 6 1 32 1 0 3/18/1991 4/17/1991 7 7 7 36 1 1 11/11/1989 6/11/1990 8 8 9 31 1 1 11/25/1989 8/25/1990 9 9 3 48 0 1 2/11/1991 5/13/1991 10 10 12 47 0 1 8/11/1989 8/11/1990 ...........(part of output omitted here)........... 91 91 4 35 0 1 4/10/1990 8/9/1990 92 92 57 36 0 1 12/11/1990 9/9/1995 93 93 1 41 1 1 12/15/1990 1/14/1991 94 94 12 36 1 0 1/13/1989 1/13/1990 95 95 7 35 1 1 8/22/1991 3/21/1992 96 96 1 34 1 1 8/2/1991 9/1/1991 97 97 5 28 0 1 5/22/1991 10/21/1991 98 98 60 29 0 0 4/2/1990 4/1/1995 99 99 2 35 1 0 5/1/1991 6/30/1991 100 100 1 34 1 1 5/11/1989 6/10/1989
Figure 1.1 on page 6 using the hmohiv data set. To control the type of symbol, a variable called psymbol is created. It takes value 1 and 2, so the symbol type will be 1 and 2.
psymbol<-censor+1
table(psymbol)
psymbol
1 2
20 80
plot(age, time, pch=(psymbol))
legend(40, 60, c("Censor=1", "Censor=0"), pch=(psymbol))
Figure 1.2 on page 7 using the hmohiv data set.
age1<-1000/age
plot(age1, time, pch=(psymbol))
legend(40, 30, c("Censor=1", "Censor=0"), pch=(psymbol))
Table 1.2 on page 14 using the data set hmohiv. Package "survival" is needed for this analysis and for most of the analyses in the book.
library(survival)
test <- survreg( Surv(time, censor) ~ age, dist="exponential")
summary(test)
Call:
survreg(formula = Surv(time, censor) ~ age, dist = "exponential")
Value Std. Error z p
(Intercept) 5.859 0.5853 10.01 1.37e-23
age -0.094 0.0158 -5.96 2.59e-09
Scale fixed at 1
Exponential distribution
Loglik(model)= -275 Loglik(intercept only)= -292.3
Chisq= 34.5 on 1 degrees of freedom, p= 4.3e-09
Number of Newton-Raphson Iterations: 4
n= 100
Figure 1.3 on page 16 using data set hmohiv and the model created for Table 1.2 in previous example.
pred <- predict(test, type="response")
ord<-order(age)
age_ord<-age[ord]
pred_ord<-pred[ord]
plot(age, time, pch=(psymbol))
lines(age_ord, pred_ord)
legend(40, 60, c("Censor=1", "Censor=0"), pch=(psymbol))
UCLA Researchers are invited to our Statistical Consulting Services
We recommend others to our list of Other Resources for Statistical Computing Help
These pages are Copyrighted (c) by UCLA Academic Technology Services
