doseInt: Fits dose interval
In lixiaomao/personalized-dosing-interval: Determining Personalized Dosing Interval

Description Usage Arguments Examples

This function is the wrapper of fitting functions for doseInt

doseInt(y = NULL, x = NULL, a = NULL, s = NULL, propensity = NULL,
  train = NULL, type = "PDI", eps = NULL, family = c("continuous",
  "ordinal", "as.ordinal"), two.sided = FALSE, method = c("rq", "svmLinear",
  "svmRadial", "RLT", "tree", "completeLinear", "completeKernel"),
  pred0 = NULL, maxiter = 20, step = 1, trace = 0, lower = TRUE,
  lambda = NULL, cost = 1, embed.mtry = 1/3, K = 10, global = F,
  margin = 0, breaks = "quantile", ...)

`train`	the training data set.
`type`	indicates whether PDI or EDI is to be calculated.
`family`	specifis the methods. 'continuous' leads to DC-algorithm,'ordinal' handles the ordinal treatments, while 'as.ordinal' cuts continuous treatment into several ordinal levels and apply the ordinal algorithm.
`two.sided`	indicator of whether two-sided interval is considered.
`method`	specified methods used for DC and ordinal approaches.
`pred0`	the initial prediction for training data set, if NULL, default initialization is used
`maxiter`	maximal iterations for DC algorithm
`step`	stepsize for DC algorithm
`trace`	the level of detail to be printed: 1,2,3,4
`lower`	True or False, Whether lower boundary is considered when applied to one-sided interval
`K`	level of ordinal treatment, default 10
`global`	if global data are used for ordinal approach, default False
`margin`	how large is the margin if global=False, default 0
`breaks`	ways 'as.ordinal' cuts continuous treatment into several ordinal levels, default 'quantile'. User can also choose 'uniform' or specify customized breaks
`test`	the testing data set. If test is NULL, use train as test.
`nfolds`	the number of folds used in cross-validation.
`alpha`	the guaranteed probability in PDI and the relative loss in EDI. Default is c(0.5,0.5).
`Eps`	parameter for the relaxation
`Lambda`	penalty for the quantile regression
`Cost`	cost for the support vector machine
`Embed.mtry`	proportion of mtry

train1=Scenario2.continuous(2000,50,1)
test1=Scenario2.continuous(10000,50,1)
fit=doseInt(train=train1,
            alpha=c(0.5,0.5),type='PDI',
            family='continuous',
            two.sided=FALSE,
            method=c('rq','svmRadial'),
            maxiter=20,step=1,trace=3,lower=TRUE,lambda=0.000001,
            K=10,global=F,margin=0)
pred=predict(fit,test1)