addSamples: Sensitivity Analysis for Binary Missing Data

Description

This is a generic method which adds bootstrap samples together with their associated results to an object of class salbm or salbmM

Usage

addSamples(obj, NBootstraps = 0, bBS = 1, nseeds = c(5,9), nseeds2 = c(-4,-5), 
                 returnJP = TRUE, returnSamples = FALSE, ...)

Arguments

obj

an object of class salbm or salbmM

NBootstraps

number of bootstraps requested

bBS

beginning index of bootstraps

nseeds

a numeric vector of length 2 to be used in makeing subsamples

nseeds2

a numeric vector of length 2 passed to randomForestSRC

returnJP

a logical, if true, the item, jps is copied to the output object

returnSamples

a logical, if true, the bootstrap samples are added to the output object

...

additional parameters passed to methods

Value

salbm returns a list which contains the following:

Main1R

results for treatment group 1 in wide format

Main1RL

results for treatment group 1 in long format

Main1wts

means and standard deviations for trt1

jps1

joint distribution returned from randomForestRSC, trt 1

Samp1R

results for bootstrap samples trt1 in wide format

Samp1RL

results for bootstrap samples trt1 in long format

Samp1wts

means and standard deviations of bootstrap samples trt1.

Main2R

results for treatment group 2 in wide format

Main2RL

results for treatment group 2 in long format

Main2wts

means and standard deviations for trt2

jps2

joint distribution returned from randomForestRSC trt 2

Samp2R

results for bootstrap samples trt2 in wide format

Samp2RL

results for bootstrap samples trt2 in long format

Samp2wts

means and standard deviations of bootstrap samples trt2.

data

the salbm data object supplied in the call to salbm

the value of K supplied in the call to salbm

ntree

the value of ntree supplied in the call to salbm

alphas

the value of alphas supplied in the call to salbm

seeds

the value of seeds supplied in the call to salbm

seeds2

the value of seeds2 supplied in the call to salbm

bBS

the value of bBS supplied in the call to salbm

eBS

the value of eBS supplied in the call to salbm

NBootstraps

the value of NBootstraps supplied in the call to salbm

Details

If the input object, obj, is of type salbm, then obj will contain an items jps1 and jps2 which are the joint probability distributions of Y_1, ..., Y_K derived from the runs of randomForestSRC applied to treatment 1 and treatment 2 respecctively. If the input object, obj, is of type salbmM, then obj will contain an items jps1 and jps2 which are a series of probability distributions derived from randomForestSRC under the Markovian assumption paramaterized by the integer m.

In either case, these jps1 and jps2 are used to create NBootstrap samples for treatment 1 and treatment 2 respectively.. Each bootstrap sample is analysed using either salbm or salbmM.

All results in the original object, obj, are copied to the output object except that any new bootstrap results are appended to the original bootstrap results.

The parameter NBootstraps determines the number of bootstraps to create. The sBS parameter allows the new bootstrap results to be be indexed by sBS:(sBS+NBootstraps-1). This is useful when running addSamples in parallel.

nseeds is of the form c(sd1,sd2). set.seed(sd1) is run once before bootstraps from treatment group 1 are created and analyzed and set.seed(sd2) is run once before bootstraps from treatment group 2 are creaded and analyzed.

nseeds2 is also of the form c(sd1,sd2). The values of sd1 and sd2 should be negative. The index for a given bootstrap sample (a value between sBS and sBS+NBootstraps-1) is subtracted from sd1 and passed to randomForestSRC when a bootstrap from treatment group 1 is analyzed and is subtraced from sd2 and passed to randomForestSRC when a bootstrap from treatment group 2 is analyzed.

The returnJP controls whether jp1 and jp2 are copied to the output object. Since these can be large, it is often advisable to set returnJP = FALSE.

Examples

Run this code

# NOT RUN {
  # Clinical trial data with two arms placebo/dose.
 
  data <- list( trt1 = trt1, trt2 = trt2 )
  # original run - no bootstraps
  R <- salbm( data = data , K = 6, ntree = 1000, 
              seeds = c(22,18), seeds2 = c(-2,-3),
              alphas = -5:5, NBootstraps=0 )

  # add 100 bootstraps
  Rupd <- addSamples(R, NBootstraps=100, sBS=1,
            nseeds = c(81,80), nseeds2 = c(-6,-1),
            returnJP=FALSE)


  # Markov assumption m = 2 again no bootstraps
  RM <- salbmM( data = data , K = 6, m = 2, 
                ntree = 1000, seeds2 = c(-2,-3),
                alphas = -5:5, NBootstraps=0 )

  # add 100 bootstraps
  RMupd <- addSamples(RM, NBootstraps=100, sBS=1,
           nseeds = c(81,80), nseeds2 = c(-6,-1),
           returnJP=FALSE)

  # Markov assumption m = 3, empirical estimates 
  RME <- salbmMEst( data = data , K = 11, m = 3, 
                MEst = TRUE, NMest = 25000,
                ntree = 1000, seeds2 = c(-2,-3),
                alphas = -5:5, NBootstraps=100 )

  # and add 100 bootstraps
  RMEupd <- addSamples(RME, NBootstraps=100, sBS=101,
           nseeds = c(81,80), nseeds2 = c(-6,-1),
           returnJP=FALSE)
# }