stsubpop-class {stepp}

Class "stsubpop"

Description

This is the S4 class for stepp subpopulation object. The subpopulations are generated based on the stepp windows and the covariate of interest.

Objects from the Class

Objects can be created by calls of the form new("stsubpop") or the constructor method stepp.subpop.

Slots

win:

Object of class "stwin"
the stepp window set up for the analysis

covar:

Object of class "numeric"
the covariate of interest

nsubpop:

Object of class "numeric"
the number of subpopulations generated

subpop:

Object of class "ANY"
a matrix of subpopulations generated based on the stepp window and the specified covariate of interest

npatsub:

Object of class "numeric" a vector of size of each subpopulation

medianz:

Object of class "numeric"
a vector of median value of the covariate of interest for each subpopulation

minc:

Object of class "numeric"
a vector of the minimum value of the covariate of interest for each subpopulation

maxc:

Object of class "numeric"
a vector of the maximum value of the covariate of interest for each subpopulation

neventsubTrt0:

Object of class "numeric" or NULL
a vector containing the number of events in each subpopulation for the baseline treatment group

neventsubTrt1:

Object of class "numeric" or NULL
a vector containing the number of events in each subpopulation for the active treatment group

init:

Object of class "logical"
a logical value indicating if the subpopulations have already been generated or not

Methods

generate

signature(.Object = "stsubpop", win, covariate, coltype, coltrt, trts, minsubpops):
a method to generate the subpopulations based on the stepp window object and the specified covariate of interest. For event-based windows, also the event type (coltype), treatment indicator (coltrt), treatments list (trts) and minimum number of subpopulations (minsubpops) must be provided

summary

signature(.Object = "stsubpop"):
a method to display the summary of the subpopulations generated

Author(s)

Wai-Ki Yip

See Also

stwin, stmodelKM, stmodelCI, stmodelGLM, steppes, stmodel, stepp.win, stepp.subpop, stepp.KM, stepp.CI, stepp.GLM, stepp.test, estimate, generate

Examples

showClass("stsubpop")

# create a steppp window
win1 <- stepp.win(type="sliding", r1=5,r2=10)

# generate the covariate of interest
Y <- rnorm(100)

# create and generate the stepp subpopulation
sp <- new("stsubpop")
sp <- generate(sp, win=win1, cov=Y)
summary(sp)

# event-based windows using the BIG data set
data(bigKM)
rxgroup <- bigKM$trt
time    <- bigKM$time
evt     <- bigKM$event
cov     <- bigKM$ki67

swin_e <- new("stwin", type = "sliding_events", e1 = 10, e2 = 20)
subp_e <- new("stsubpop")
subp_e <- generate(subp_e, win = swin_e, covariate = cov, coltype = evt,
                   coltrt = rxgroup, trts = c(1, 2), minsubpops = 5)
summary(subp_e)