DubRentVariable

library(caret)

# ??createDataPartition
# ?train
# ?expand.grid
# ??varImp showcases variable importance of the variables used in the final model.
# ??defaultSummary

#use the read.csv function to read the data from a CSV file
#drl_cpi_prr_vr_housingstock_0.csv
#into a data frame
df <- read.csv(file.choose(), header=TRUE)

# Data Loading and Splitting
# 80% train data
trainIndex <- createDataPartition(df$Rent, p = .8,
                                  list = FALSE,
                                  times = 1)

head(trainIndex)

dfTrain <- df[ trainIndex,]
dfTest  <- df[-trainIndex,]

# Model Building and Tuning
lmFit<-train(Rent~., data = dfTrain, method = "lm")
summary(lmFit)

ctrl<-trainControl(method = "cv",number = 10)

lmCVFit<-train(Rent ~ ., data = df, method = "lm", trControl = ctrl, metric="Rsquared")

summary(lmCVFit)

# Model Diagnostics and Scoring
residuals<-resid(lmFit)

predictedValues<-predict(lmFit)

plot(dfTrain$Rent,residuals)

abline(0,0)

plot(dfTrain$Rent,predictedValues)

varImp(lmFit)

plot(varImp(lmFit))

predictedVal<-predict(lmFit,val)

#
modelvalues<-data.frame(obs = dfTrain$Rent, pred=predictedVal)

defaultSummary(modelvalues)