diff --git a/R/plotCGains.R b/R/plotCGains.R
index 7f270d9e..54bf4d55 100644
--- a/R/plotCGains.R
+++ b/R/plotCGains.R
@@ -3,8 +3,6 @@
 #' @description Cumulative Gains Chartis a plot of the rate of positive prediction against true positive rate for the different thresholds.
 #' It is useful for measuring and comparing the accuracy of the classificators.
 #' @param object An object of class ModelAudit
-#' @param newdata optionally, a data frame in which to look for variables with which to plot CGains curve. If omitted, the data used to build model will be used.
-#' @param newy optionally, required if newdata used. Response vector for new data.
 #' @param ... other modelAudit objects to be plotted together
 #'
 #' @return ggplot object
@@ -14,33 +12,20 @@
 #' @import ggplot2
 #' @import ROCR
 #'
-#' @examples
-#' library(auditor)
-#' library(mlbench)
-#' library(randomForest)
-#' data("PimaIndiansDiabetes")
-#'
-#' model_rf <- randomForest(diabetes~., data=PimaIndiansDiabetes)
-#' au_rf <- audit(model_rf, label="rf")
-#' plotCGains(au_rf)
-#'
-#' model_glm <- glm(diabetes~., family=binomial,	data=PimaIndiansDiabetes)
-#' au_glm <- audit(model_glm)
-#' plotCGains(au_rf, au_glm)
 #'
 #' @export
 
 
-plotCGains <- function(object, ..., newdata = NULL, newy){
+plotCGains <- function(object, ...){
   if(class(object)!="modelAudit") stop("plotCGains requires object class modelAudit.")
   rpp <- tpr <- label <- NULL
-  df <- getCGainsDF(object, newdata, newy)
+  df <- getCGainsDF(object)
 
   dfl <- list(...)
   if (length(dfl) > 0) {
     for (resp in dfl) {
       if(class(resp)=="modelAudit"){
-        df <- rbind( df, getCGainsDF(resp, newdata, newy) )
+        df <- rbind( df, getCGainsDF(resp) )
       }
     }
   }
@@ -52,15 +37,10 @@ plotCGains <- function(object, ..., newdata = NULL, newy){
     theme_light()
 }
 
-getCGainsDF <- function(object, newdata, newy){
-  if (is.null(newdata)) {
-    predictions <-   object$fitted.values
-    y <- object$y
-  } else {
-    if(is.null(newy)) stop("newy must be provided.")
-    predictions <- object$predict.function(object$model, newdata)
-    y <- newy
-  }
+getCGainsDF <- function(object){
+
+  predictions <- object$fitted.values
+  y <- as.numeric(as.character(object$y))
 
   pred <- prediction(predictions, y)
   gain <- performance(pred, "tpr", "rpp")
diff --git a/R/plotLift.R b/R/plotLift.R
index e12b8778..1b575800 100644
--- a/R/plotLift.R
+++ b/R/plotLift.R
@@ -45,7 +45,7 @@ plotLIFT <- function(object, ..., groups = 10, cumulative = TRUE){
 
 getLIFTDF <- function(object, n.groups, cumulative = TRUE){
   pred <- NULL
-  y = object$y
+  y = as.numeric(as.character(object$y))
   df <- data.frame(pred=object$fitted.values, y=y)
   df <- arrange(df, desc(pred))
 
diff --git a/R/plotModelPCA.R b/R/plotModelPCA.R
index 630448bb..7dc81f23 100644
--- a/R/plotModelPCA.R
+++ b/R/plotModelPCA.R
@@ -29,7 +29,7 @@ plotModelPCA <- function(object, ...){
     }
   }
 
-  res.pca <- prcomp(df, scale = TRUE)
+  res.pca <- prcomp(df, scale = FALSE)
 
   fviz_pca_biplot(res.pca,
                   repel = TRUE,
diff --git a/man/plotCGains.Rd b/man/plotCGains.Rd
index 452e6edc..59cff486 100644
--- a/man/plotCGains.Rd
+++ b/man/plotCGains.Rd
@@ -4,16 +4,12 @@
 \alias{plotCGains}
 \title{Cumulative Gains Chart}
 \usage{
-plotCGains(object, ..., newdata = NULL, newy)
+plotCGains(object, ...)
 }
 \arguments{
 \item{object}{An object of class ModelAudit}
 
 \item{...}{other modelAudit objects to be plotted together}
-
-\item{newdata}{optionally, a data frame in which to look for variables with which to plot CGains curve. If omitted, the data used to build model will be used.}
-
-\item{newy}{optionally, required if newdata used. Response vector for new data.}
 }
 \value{
 ggplot object
@@ -21,21 +17,6 @@ ggplot object
 \description{
 Cumulative Gains Chartis a plot of the rate of positive prediction against true positive rate for the different thresholds.
 It is useful for measuring and comparing the accuracy of the classificators.
-}
-\examples{
-library(auditor)
-library(mlbench)
-library(randomForest)
-data("PimaIndiansDiabetes")
-
-model_rf <- randomForest(diabetes~., data=PimaIndiansDiabetes)
-au_rf <- audit(model_rf, label="rf")
-plotCGains(au_rf)
-
-model_glm <- glm(diabetes~., family=binomial,	data=PimaIndiansDiabetes)
-au_glm <- audit(model_glm)
-plotCGains(au_rf, au_glm)
-
 }
 \seealso{
 \code{\link{plot.modelAudit}}