DynTxRegime (version 3.2)

.newTxInfo: Create Treatment Information Objects

Description

Given a definition for feasible sets and treatment name, creates an appropriately classed treatment information object. Method is not exported.

Usage

.newTxInfo(fSet, txName, ...)
# S4 method for NULL,character
.newTxInfo(fSet, txName, data, suppress, verify = TRUE)
# S4 method for function,character
.newTxInfo(fSet, txName, data, suppress, verify = TRUE)
# S4 method for list,list
.newTxInfo(fSet, txName, data, suppress, verify = TRUE)
# S4 method for NULL,list
.newTxInfo(fSet, txName, data, suppress, verify = TRUE)

Arguments

fSet

NULL, "function", or "list" of "function"s specifying subsets.

txName

The column header of "data.frame" indicating the treatment variable. If list, column header of "data.frame" indicating treatment at each decision point.

...

Used to pass arguments that are required but the class of which does not determine the method that is selected.

data

"data.frame" of covariates and treatment histories.

suppress

T/F indicating if screen prints are generated.

verify

T/F indicating if treatments should be verified.

Examples

# NOT RUN {
data(bmiData)
bmiData$A1 <- as.factor(bmiData$A1)
bmiData$A2 <- as.factor(bmiData$A2)
bmiData$A3 <- 0L
bmiData$A3[bmiData$A1 == "MR"] <- 1L
bmiData$A4 <- 0L
bmiData$A5 <- as.factor("CD")

# Integer treatment with no subsetting
obj <- DynTxRegime:::.newTxInfo(fSet = NULL, 
                                txName = "A3", 
                                data = bmiData,  
                                suppress = TRUE,  
                                verify = TRUE)
is(obj)

# Integer treatment with only 1 tx in data
# Illegal if verification is requested
obj <- try(DynTxRegime:::.newTxInfo(fSet = NULL,  
                                    txName = "A4",  
                                    data = bmiData,  
                                    suppress = TRUE,  
                                    verify = TRUE))

# Integer treatment with only 1 tx in data
# legal if verification is turned off
obj <- DynTxRegime:::.newTxInfo(fSet = NULL,  
                                txName = "A4",  
                                data = bmiData,  
                                suppress = TRUE,  
                                verify = FALSE)
is(obj)

# Factor treatment with no subsetting
obj <- DynTxRegime:::.newTxInfo(fSet = NULL,  
                                txName = "A2",  
                                data = bmiData,  
                                suppress = TRUE,  
                                verify = TRUE)
is(obj)

# Factor treatment with only 1 tx in data
# Illegal if verification is requested
obj <- try(DynTxRegime:::.newTxInfo(fSet = NULL,  
                                txName = "A5",  
                                data = bmiData,  
                                suppress = TRUE,  
                                verify = TRUE))

# Factor treatment with only 1 tx in data
# legal if verification is turned off
obj <- DynTxRegime:::.newTxInfo(fSet = NULL,  
                                txName = "A5",  
                                data = bmiData,  
                                suppress = TRUE,  
                                verify = FALSE)
is(obj)

fSet1 <- function(data){
           subsets <- list(list("subset1",c(0L,1L)),
                           list("subset2",c(0L,1L)))
           txOpts <- character(nrow(data))
           txOpts[data$A1 == "CD"] <- "subset1"
           txOpts[data$A1 == "MR"] <- "subset2"
           return(list("subsets" = subsets, "txOpts" = txOpts))
         }

# Integer treatment with subsetting
obj <- DynTxRegime:::.newTxInfo(fSet = fSet1, 
                                txName = "A3", 
                                data = bmiData, 
                                suppress = TRUE, 
                                verify = TRUE)
is(obj)

# Integer treatment with subsetting only 1 tx in data
# Illegal if verification is on
obj <- try(DynTxRegime:::.newTxInfo(fSet = fSet1, 
                                    txName = "A4", 
                                    data = bmiData, 
                                    suppress = TRUE, 
                                    verify = TRUE))

# Integer treatment with subsetting only 1 tx in  data
# Legal if verification is off
obj <- DynTxRegime:::.newTxInfo(fSet = fSet1, 
                                txName = "A4", 
                                data = bmiData, 
                                suppress = TRUE, 
                                verify = FALSE)
is(obj)

fSet2 <- function(data){
           subsets <- list(list("subset1", c("CD","MR")),
                           list("subset2", c("CD","MR")))
          txOpts <- character(nrow(data))
          txOpts[data$A1 == "CD"] <- "subset1"
          txOpts[data$A1 == "MR"] <- "subset2"
          return(list("subsets" = subsets, "txOpts" = txOpts))
         }

# Factor treatment with subsetting
obj <- DynTxRegime:::.newTxInfo(fSet = fSet2, 
                                txName = "A2", 
                                data = bmiData, 
                                suppress = TRUE, 
                                verify = TRUE)
is(obj)

# Factor treatment with subsetting only 1 tx in data
# Illegal of verification is on
obj <- try(DynTxRegime:::.newTxInfo(fSet = fSet2, 
                                txName = "A5", 
                                data = bmiData, 
                                suppress = TRUE, 
                                verify = TRUE))

# Factor treatment with subsetting only 1 tx in data
# Legal of verification is off
obj <- DynTxRegime:::.newTxInfo(fSet = fSet2, 
                                txName = "A5", 
                                data = bmiData, 
                                suppress = TRUE, 
                                verify = FALSE)
is(obj)

# Multiple decision points no subsetting
obj <- DynTxRegime:::.newTxInfo(fSet = NULL, 
                                txName = list("A1","A2"), 
                                data  = bmiData, 
                                suppress = TRUE, 
                                verify = TRUE)
is(obj)
is(obj[[1L]])
is(obj[[2L]])

# Multiple decision points no subsetting
# Only 1 tx in data
# Legal if verification is turned off
obj <- DynTxRegime:::.newTxInfo(fSet = NULL, 
                                txName = list("A1","A4"), 
                                data  = bmiData, 
                                suppress = TRUE, 
                                verify = FALSE)
is(obj)
is(obj[[1L]])
is(obj[[2L]])

# Multiple decision points no subsetting
# Only 1 tx in data
# Illegal if verification is turned on
obj <- try(DynTxRegime:::.newTxInfo(fSet = NULL, 
                                    txName = list("A1","A4"), 
                                    data  = bmiData, 
                                    suppress = TRUE, 
                                    verify = TRUE))

# Multiple decision points with subsetting
obj <- DynTxRegime:::.newTxInfo(fSet = list(fSet1,fSet2), 
                                txName = list("A3","A2"), 
                                data  = bmiData, 
                                suppress = TRUE, 
                                verify = TRUE)
is(obj)
is(obj[[1L]])
is(obj[[2L]])

# Multiple decision points with subsetting
# Only 1 tx in data
# Legal if verification is turned off
obj <- DynTxRegime:::.newTxInfo(fSet = list(fSet1,fSet2), 
                                txName = list("A1","A5"), 
                                data  = bmiData, 
                                suppress = TRUE, 
                                verify = FALSE)
is(obj)
is(obj[[1L]])
is(obj[[2L]])

# Multiple decision points with subsetting
# Only 1 tx in data
# Illegal if verification is turned on
obj <- try(DynTxRegime:::.newTxInfo(fSet = list(fSet1,fSet2), 
                                    txName = list("A1","A5"), 
                                    data  = bmiData, 
                                    suppress = TRUE, 
                                    verify = TRUE))
# }