fitDRM: fitting dose-response model according to the identified pattern.

Description

function to fit several dose-response candidate models according to the identified pattern, and combine their results using model selection and/or model averaging.

Usage

fitDRM(inputDataset, dose, response, ID, subsettingID = NULL,
  transform = c("none", "log", "sRoot", "qRoot", "boxcox"),
  addCovars = ~1, patternClusters, EDp = 0.5, addCovarsVar = TRUE,
  alpha = 0.05, na.rm = FALSE, imputationMethod = c("mean",
  "median"), nCores = 1)

Arguments

inputDataset

a data frame containing the input dataset, it should at least include dose, response, and ID

dose

either a single string or a scalar, indicating the name of the dose column or its index.

response

either a single string or a scalar, indicating the name of the response column or its index.

either a single string or a scalar, indicating the name of the ID column or its index.

subsettingID

a vector of ID's of the subjects, in case one wants to fit the models only to a subset of the data. Default is NULL, i.e., all the subjects in the inputDataset will be used.

transform

single string indicating what kind of transform should be applied on the response data. It takes "none" (no transform, dafault), "log" (natural log), "sRoot (square root), and "qRoot" (cubic root), and "boxcox" (Box-Cox transformation).

addCovars

formula specifying extra linear covariate, e.g., ~x1+x2

patternClusters

a vector of the same length as the number of rows in inputData (number of subjects) indicating a pattern for each subject. Note that the keywords which are recognized are: "increasing", "decreasing", "flat", "complete", and "up down max at x" and "down up min at x", which x is one of the doses. The "flat" and "complete" patterns would not be considered.

EDp

scalar in (0,1), indicatign with EDp should be computed, default is 0.5 (ED50).

addCovarsVar

logical variable (TRUE as default), idicating whether the variance of the extra covariates preesented in addCovars (unless it is only intercept) should also be computed or not.

alpha

scalar in (0,1), level of significance with default alpha = 0.05.

na.rm

logical variable indicatign whether missing values should be removed (TRUE) or not (FALSE, default)

imputationMethod

signle string taking calues from "mean" (default), and "median", which indicates how the missing values should be treated. "mean" would replace them with the mean of the observed ones, and "median" will use median of them for imputation.

nCores

scalar, indicating the number of cores should be used to perform LRT and MCT tests. Default is 1 which means sequantial computation (no prallel computation).

Value

an object fo class fittedDRM which is a list with the following objects: fittedModels the outcome of DoseFinding::fitMod for all the suitable models estAICNonmonotone: the computed AIC for the models fitted to the subjects with a non-monotone pattern estEDpNonmonotone: the computed EDp for the models fitted to the subjects with a non-monotone pattern estAICMonmonotone: the computed AIC for the models fitted to the subjects with a monotone pattern estEDpMonmonotone: the computed EDp for the models fitted to the subjects with a monotone pattern extraCovarsMonotone: if any extra covariates are added to the model their estimates and possibly standard errors (if addCovarsVar = TRUE) are gievn for subjects with monotone pattern. extraCovarsNonmonotone: if any extra covariates are added to the model their estimates and possibly standard errors (if addCovarsVar = TRUE) are gievn for subjects with non-monotone pattern.

Details

Note that the dose column of the inputDataset should be a numeric variable.

Examples

Run this code

# NOT RUN {
## gnerating data
set.seed(11)
doses2Use <-  c(0, 5, 20)
numRep2Use <- c(6, 3, 3)
generatedData <- cbind(rep(1,sum(numRep2Use)),
MCPMod::genDFdata("logistic",c(5, 3, 10, 0.05), doses2Use, 
numRep2Use, 1), 
		matrix(rnorm(1*sum(numRep2Use)), sum(numRep2Use), 1))
colnames(generatedData) <- c("ID", "dose", "response", "x1")
for (iGen in 2:15){
	genData0 <- cbind(rep(iGen,sum(numRep2Use)),
MCPMod::genDFdata("logistic",c(5, 3, 10, 0.05), doses2Use, 
numRep2Use, 1), matrix(rnorm(1*sum(numRep2Use)), 
sum(numRep2Use), 1))
	colnames(genData0) <- c("ID", "dose", "response", "x1")
	generatedData <- rbind(generatedData, genData0)
}
## transforming it for clustering
toInput <- inputDataMaker(2, 3, 1, generatedData)
## general pattern clustering
generalPatternClust <- generalPatternClustering(
inputData = toInput$inputData, colsData = toInput$colsData ,
colID = toInput$colID, doseLevels = toInput$doseLevels, 
numReplications = toInput$numReplicates, na.rm = FALSE, 
imputationMethod = "mean", ORICC = "two", transform = "none",
plotFormat = "eps", LRT = TRUE, MCT = TRUE,
		adjustMethod = "BH", nPermute = 100, useSeed = NULL, 
theLeastNumberOfMethods = 2, alpha = 0.05, nCores = 1)
## fitDRM 
fittedModel <- fitDRM (inputDataset = generatedData, dose = 2, 
response = 3, ID = 1, subsettingID = NULL, 
		transform = c("none"), addCovars = ~x1, 
		patternClusters = 
generalPatternClust$clusteringORICC2Results$clusteringResultsORICC2, 
		EDp = 0.5, addCovarsVar = TRUE, alpha = 0.05, na.rm = FALSE, 
imputationMethod = c("mean"), nCores = 1)

# }