Model Automation in R

model-automation-options.R

#http://archive.ics.uci.edu/ml/datasets/Bank+Marketing
data <- read.csv("~/in/bank/bank-full.csv",header=T,
         sep=";")

####Decision Tree####
library(rpart)
library(rpart.plot)

rp <- rpart(y~., data = data)
rpart.plot(rp)

####Random Forest Models####
library(randomForest)
rf <- randomForest(y~., data = data)
rf$importance #Var Name + Importance
varImpPlot(rf) #Visualization

####Step Wise Regression####
data("mtcars")
mt <- mtcars
#Get the Independent Variables
#(and exclude hp dependent variable)
indep_vars <-paste(names(mt)[-which(names(mt)=="hp")],
                   collapse="+")
#Turn those variable names into a formula
upper_form = formula(paste("~",indep_vars,collapse=""))
#~mpg + cyl + disp + drat + wt + qsec + vs + am + gear + carb

library(MASS)
mod <- lm(hp~.,data=mt)
#Step Backward and remove one variable at a time
stepAIC(mod,direction = "backward",trace = T)
#Create a model using only the intercept
mod_lower = lm(hp~1,data=mt)
#Step Forward and add one variable at a time
stepAIC(mod_lower,direction = "forward",
        scope=list(upper=upper_form,lower=~1))
#Step Forward or Backward each step starting with a intercept model
stepAIC(mod_lower,direction = "both",
        scope=list(upper=upper_form,lower=~1))

####Auto Arima####
library(forecast)
library(fpp)
#Step Backward and remove one variable at a time
data("elecequip")
ee <- elecequip[1:180]
model <- auto.arima(ee,stationary = T)

plot(forecast(model,h=10))
lines(x = 181:191, y= elecequip[181:191],
      type = 'l', col = 'red')

####Hyper Parameter Search ####
library(caret)
#Step Backward and remove one variable at a time
tctrl <- trainControl(method = "cv",number=10,
                      repeats=10)
rpart_opts <- expand.grid(cp = seq(0.0,0.01, by = 0.001))

rpart_model <- train(y~., data = data, method="rpart", 
                     metric = "Kappa", trControl = tctrl, 
                     tuneGrid = rpart_opts)