version 1.3.0

DESCRIPTION

@@ -2,19 +2,19 @@ Package: gesttools
 Type: Package
 Title: General Purpose G-Estimation for End of Study or Time-Varying
         Outcomes
-Version: 1.2.0
+Version: 1.3.0
 Author: Daniel Tompsett, Stijn Vansteelandt, Oliver Dukes, Bianca De Stavola
 Maintainer: Daniel Tompsett <danieltompsettwork@gmail.com>
 Description: Provides a series of general purpose tools to perform g-estimation using the methods described in Sjolander and Vansteelandt (2016) <doi:10.1515/em-2015-0005> and Dukes and Vansteelandt <doi:10.1093/aje/kwx347>. The package allows for g-estimation in a wide variety of circumstances, including an end of study or time-varying outcome, and an exposure that is a binary, continuous, or a categorical variable with three or more categories. The package also supports g-estimation with time-varying causal effects and effect modification by a confounding variable.
 License: GPL-3
 Encoding: UTF-8
-LazyData: false
-RoxygenNote: 7.1.1
+RoxygenNote: 7.1.2
 Imports: DataCombine, tidyr, tibble, tidyselect, geeM, rsample, nnet,
-        magrittr
+        magrittr, testthat
 URL: https://github.com/danieltompsett/gesttools
 BugReports: https://github.com/danieltompsett/gesttools/issues
+Config/testthat/edition: 3
 NeedsCompilation: no
-Packaged: 2021-06-10 12:12:03 UTC; danie
+Packaged: 2021-12-28 22:44:31 UTC; danie
 Repository: CRAN
-Date/Publication: 2021-06-10 12:30:02 UTC
+Date/Publication: 2022-01-04 09:20:02 UTC

MD5

@@ -1,12 +1,14 @@
-21dc5399dd23c8e40c36d14612a5e92c *DESCRIPTION
-85277f9fbbc040992f797e1d0aee8be1 *NAMESPACE
-957e569bfd4f200d7a3ac3a4aed457da *R/FormatData.R
-eec3b250c6b064fe2e9ff97189bee7d3 *R/dataexamples.R
-ba98b87a6305b3d023fc9a14e01362e9 *R/gestMultiple.R
-a2a9bc0617e5df5dc05692e2dc917354 *R/gestSingle.R
-bd338af86889cb644385ab93487954b4 *R/gestboot.R
-f4895791aad931c909d94ffdede6d515 *man/FormatData.Rd
-5cdc73c2a35dd4622385126619addb7f *man/dataexamples.Rd
-8b8627228f71f64f80a7769077c8533d *man/gestMultiple.Rd
-df0fee5fd62ac25d53aeb768385935ac *man/gestSingle.Rd
-2fc955afa4df12e21ceb79fec5f9fe33 *man/gestboot.Rd
+b82155154637ea1fa04f26312d1170e9 *DESCRIPTION
+78eb1179edf4cb26d0a16ca27ff9bd65 *NAMESPACE
+411ade2a445c6b26e8f6f50ca18224a0 *R/FormatData.R
+38bec68bc8fb55194b76f15352bae11b *R/dataexamples.R
+353ac2dbddb7c64379a2aa90456f2637 *R/gestMultiple.R
+b9dda3b5edda76a02597ae78ee92a5d8 *R/gestSingle.R
+8b6dbd00b05710cbafa72d0b1af2238b *R/gestboot.R
+d3026cc0204b91ad959499d1c45f87c7 *man/FormatData.Rd
+b94e5d1649b7323b4f9036ce39b39a59 *man/dataexamples.Rd
+5c03f475fe3ae6f245d0ab2ea45c1e60 *man/gestMultiple.Rd
+2b2e24a9cefda5abd6ef53156d4941a4 *man/gestSingle.Rd
+dcf4c2fb2f43e721e41f712897ad8003 *man/gestboot.Rd
+baadd5fcff0f63aff29edee57598e176 *tests/testthat.R
+04d67ec07380ca4555a4988b2db60c95 *tests/testthat/test.R

NAMESPACE

@@ -5,13 +5,15 @@ export(dataexamples)
 export(gestMultiple)
 export(gestSingle)
 export(gestboot)
+import(testthat)
 importFrom(DataCombine,slide)
 importFrom(geeM,geem)
 importFrom(magrittr,"%>%")
 importFrom(nnet,multinom)
 importFrom(rsample,bootstraps)
 importFrom(stats,Gamma)
 importFrom(stats,complete.cases)
+importFrom(stats,formula)
 importFrom(stats,gaussian)
 importFrom(stats,glm)
 importFrom(stats,plogis)

R/FormatData.R

@@ -22,8 +22,7 @@
 #' corresponding to the lagged histories of all variables included in \code{varying}.
 #' These will be labeled as \code{LagVari} where \code{Var} is the variable name and \code{i}
 #' indicates how much the variable is lagged by. For example \code{LagAn2} is the value of \code{An}, 2
-#' time periods prior. Note that \code{LagAn1} is not generated as this is automatically included
-#' in the g-estimation functions.
+#' time periods prior.
 #' @param GenerateHistoryMax An optional positive integer specifying \code{GenerateHistory} to generate exposure histories
 #' up to \code{GenerateHistoryMax} time periods prior.
 #'
@@ -35,62 +34,66 @@
 #' \code{as.numeric()} variable.
 #'
 #' @examples
-#' data<-dataexamples(n=1000,seed=3456,Censoring=TRUE)$datagest
-#' #To demonstrate the function we
-#' #Delete the third row, corresponding to the entry for ID 1 at time 3
-#' data<-data[-3,]
-#' datanew<-FormatData(data=data,idvar="id",timevar="time",An="A",
-#' varying=c("A","L"),GenerateHistory=FALSE,GenerateHistoryMax=NA)
+#' data <- dataexamples(n = 1000, seed = 3456, Censoring = TRUE)$datagest
+#' # To demonstrate the function we
+#' # Delete the third row, corresponding to the entry for ID 1 at time 3
+#' data <- data[-3, ]
+#' datanew <- FormatData(
+#'   data = data, idvar = "id", timevar = "time", An = "A",
+#'   Cn = "C", varying = c("A", "L"), GenerateHistory = TRUE, GenerateHistoryMax = 1
+#' )
 #' head(datanew)
-#' #Note that the missing entry has been re-added,
-#' #with missing values for A and L in the third row
-#' #An example with lagged history of time varying variables created.
-#' data<-dataexamples(n=1000,seed=3456,Censoring=TRUE)$datagestmultcat
-#' datanew<-FormatData(data=data,idvar="id",timevar="time",An="A",
-#' Cn="C",varying=c("A","L"),GenerateHistory=TRUE,GenerateHistoryMax=NA)
+#' # Note that the missing entry has been re-added,
+#' # with missing values for A and L in the third row
+#' # An example with lagged history of time varying variables created.
+#' data <- dataexamples(n = 1000, seed = 3456, Censoring = TRUE)$datagestmultcat
+#' datanew <- FormatData(
+#'   data = data, idvar = "id", timevar = "time", An = "A",
+#'   Cn = "C", varying = c("Y","A", "L"), GenerateHistory = TRUE, GenerateHistoryMax = NA
+#' )
 #' head(datanew)
 #' @export
 
 
-FormatData<-function(data,idvar,timevar,An,varying,Cn=NA,GenerateHistory=FALSE,
-                     GenerateHistoryMax=NA){
-if(!is.data.frame(data))(stop("Either no data set has been given, or it is not in a data frame."))
-if (!is.na(Cn)){
-  varying<-c(varying,Cn)
-}
-
-datwide<-reshape(data,direction="wide",timevar=timevar,idvar=idvar,v.names=varying)
-datrec<-reshape(datwide,direction="long",timevar=timevar,idvar=idvar)
-datrec<-datrec[order(datrec[,idvar],datrec[,timevar]),]
+FormatData <- function(data, idvar, timevar, An, varying, Cn = NA, GenerateHistory = FALSE,
+                       GenerateHistoryMax = NA) {
+  if (!is.data.frame(data)) (stop("Either no data set has been given, or it is not in a data frame."))
+  if (!is.na(Cn)) {
+    varying <- c(varying, Cn)
+  }
 
-if(GenerateHistory==TRUE){
-  T<-max(datrec[,timevar])
-  if (T<=1)(stop('Lagged variables cannot be created with only 1 time period'))
-  if(is.na(GenerateHistoryMax)==TRUE){
-    GenerateHistoryMax<-T-1}
-  varying<-varying[!(varying %in% Cn)]
-  histmax<-min(T-1,GenerateHistoryMax)
+  datwide <- reshape(data, direction = "wide", timevar = timevar, idvar = idvar, v.names = varying)
+  datrec <- reshape(datwide, direction = "long", timevar = timevar, idvar = idvar)
+  datrec <- datrec[order(datrec[, idvar], datrec[, timevar]), ]
 
-    #Function to generate lagged variables
-    lagged<-function(name){
-    for (i in 1:histmax){
-      datrec<-slide(data=datrec,Var=name,GroupVar="id",slideBy=-i,NewVar=paste("Lag",i,name,sep=""),reminder=F)
-      #Set Value of lagged variable that do not exists to either 0, or the reference category.
-      if(is.factor(datrec[,name])==TRUE){
-        datrec[,paste("Lag",i,name,sep="")]<-as.factor(datrec[,paste("Lag",i,name,sep="")])
-        levels(datrec[,paste("Lag",i,name,sep="")])<-levels(datrec[,name])
-        datrec[datrec[,timevar] %in% seq(1,i,by=1),paste("Lag",i,name,sep="")]<-levels(datrec[,name])[1]
+  if (GenerateHistory == TRUE) {
+    T <- max(datrec[, timevar])
+    if (T <= 1) (stop("Lagged variables cannot be created with only 1 time period"))
+    if (is.na(GenerateHistoryMax) == TRUE) {
+      GenerateHistoryMax <- T - 1
+    }
+    varying <- varying[!(varying %in% Cn)]
+    histmax <- min(T - 1, GenerateHistoryMax)
 
-      }else{
-        datrec[datrec[,timevar] %in% seq(1,i,by=1),paste("Lag",i,name,sep="")]<-0}
+    # Function to generate lagged variables
+    lagged <- function(name) {
+      for (i in 1:histmax) {
+        datrec <- slide(data = datrec, Var = name, GroupVar = "id", slideBy = -i, NewVar = paste("Lag", i, name, sep = ""), reminder = F)
+        # Set Value of lagged variable that do not exists to either 0, or the reference category.
+        if (is.factor(datrec[, name]) == TRUE) {
+          datrec[, paste("Lag", i, name, sep = "")] <- as.factor(datrec[, paste("Lag", i, name, sep = "")])
+          levels(datrec[, paste("Lag", i, name, sep = "")]) <- levels(datrec[, name])
+          datrec[datrec[, timevar] %in% seq(1, i, by = 1), paste("Lag", i, name, sep = "")] <- levels(datrec[, name])[1]
+        } else {
+          datrec[datrec[, timevar] %in% seq(1, i, by = 1), paste("Lag", i, name, sep = "")] <- 0
+        }
       }
-    return(datrec)
+      return(datrec)
     }
 
-  datrec<-Reduce(merge,lapply(varying,lagged))
-  #Remove first lagged exposure as it is already generated by functions
-  datrec<-datrec[,-which(names(datrec) %in% c(paste("Lag",1,An,sep="")))]
-}
+    datrec <- Reduce(merge, lapply(varying, lagged))
+  }
 
 
-return(datrec)}
+  return(datrec)
+}