variable_selection.Rmd

---
title: "Variable Selection Methods"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{Variable Selection Methods}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r, echo=FALSE, message=FALSE}
library(olsrr)
library(dplyr)
library(ggplot2)
library(gridExtra)
library(purrr)
library(tibble)
library(nortest)
library(goftest)
```

<style type="text/css">
#best-subset-regression pre { /* Code block */
  font-size: 10px
}
</style>

## Introduction

## All Possible Regression

All subset regression tests all possible subsets of the set of potential independent variables. If there are K potential independent variables (besides the constant), then there are $2^{k}$ distinct subsets of them to be tested. For example, if you have 10 candidate independent variables, the number of subsets to be tested is $2^{10}$, which is 1024, and if you have 20 candidate variables, the number is $2^{20}$, which is more than one million.

```{r allsub}
model <- lm(mpg ~ disp + hp + wt + qsec, data = mtcars)
ols_step_all_possible(model)
```

The `plot` method shows the panel of fit criteria for all possible regression methods. 

```{r allsubplot, fig.width=10, fig.height=10, fig.align='center'}
model <- lm(mpg ~ disp + hp + wt + qsec, data = mtcars)
k <- ols_step_all_possible(model)
plot(k)
```

## Best Subset Regression

Select the subset of predictors that do the best at meeting some well-defined objective criterion, 
such as having the largest R2 value or the smallest MSE, Mallow's Cp or AIC.

```{r bestsub, size='tiny'}
model <- lm(mpg ~ disp + hp + wt + qsec, data = mtcars)
ols_step_best_subset(model)
```

The `plot` method shows the panel of fit criteria for best subset regression methods. 

```{r bestsubplot, fig.width=10, fig.height=10, fig.align='center'}
model <- lm(mpg ~ disp + hp + wt + qsec, data = mtcars)
k <- ols_step_best_subset(model)
plot(k)
```

## Stepwise Forward Regression

Build regression model from a set of candidate predictor variables by entering predictors based on 
p values, in a stepwise manner until there is no variable left to enter any more. The model should include all the candidate predictor variables. If details is set to `TRUE`, each step is displayed.

### Variable Selection

```{r stepf1}
# stepwise forward regression
model <- lm(y ~ ., data = surgical)
ols_step_forward_p(model)
```

### Plot

```{r stepf2, fig.width=10, fig.height=10, fig.align='center'}
model <- lm(y ~ ., data = surgical)
k <- ols_step_forward_p(model)
plot(k)
```

### Detailed Output

```{r stepwisefdetails}
# stepwise forward regression
model <- lm(y ~ ., data = surgical)
ols_step_forward_p(model, details = TRUE)
```

## Stepwise Backward Regression

Build regression model from a set of candidate predictor variables by removing predictors based on 
p values, in a stepwise manner until there is no variable left to remove any more. The model should include all the candidate predictor variables. If details is set to `TRUE`, each step is displayed.

### Variable Selection

```{r stepb, fig.width=10, fig.height=10, fig.align='center'}
# stepwise backward regression
model <- lm(y ~ ., data = surgical)
ols_step_backward_p(model)
```

### Plot

```{r stepb2, fig.width=10, fig.height=10, fig.align='center'}
model <- lm(y ~ ., data = surgical)
k <- ols_step_backward_p(model)
plot(k)
```

### Detailed Output

```{r stepwisebdetails}
# stepwise backward regression
model <- lm(y ~ ., data = surgical)
ols_step_backward_p(model, details = TRUE)
```

## Stepwise Regression

Build regression model from a set of candidate predictor variables by entering and removing predictors based on 
p values, in a stepwise manner until there is no variable left to enter or remove any more. The model should include all the candidate predictor variables. If details is set to `TRUE`, each step is displayed.

### Variable Selection

```{r stepwise1}
# stepwise regression
model <- lm(y ~ ., data = surgical)
ols_step_both_p(model)
```

### Plot

```{r stepwise2, fig.width=10, fig.height=10, fig.align='center'}
model <- lm(y ~ ., data = surgical)
k <- ols_step_both_p(model)
plot(k)
```

### Detailed Output

```{r stepwisedetails}
# stepwise regression
model <- lm(y ~ ., data = surgical)
ols_step_both_p(model, details = TRUE)
```

## Stepwise AIC Forward Regression

Build regression model from a set of candidate predictor variables by entering predictors based on 
Akaike Information Criteria, in a stepwise manner until there is no variable left to enter any more.
The model should include all the candidate predictor variables. If details is set to `TRUE`, each step is displayed.

### Variable Selection

```{r stepaicf1}
# stepwise aic forward regression
model <- lm(y ~ ., data = surgical)
ols_step_forward_aic(model)
```

### Plot

```{r stepaicf2, fig.width=5, fig.height=5, fig.align='center'}
model <- lm(y ~ ., data = surgical)
k <- ols_step_forward_aic(model)
plot(k)
```

### Detailed Output

```{r stepwiseaicfdetails}
# stepwise aic forward regression
model <- lm(y ~ ., data = surgical)
ols_step_forward_aic(model, details = TRUE)
```

## Stepwise AIC Backward Regression

Build regression model from a set of candidate predictor variables by removing predictors based on 
Akaike Information Criteria, in a stepwise manner until there is no variable left to remove any more.
The model should include all the candidate predictor variables. If details is set to `TRUE`, each step is displayed.

### Variable Selection

```{r stepaicb1}
# stepwise aic backward regression
model <- lm(y ~ ., data = surgical)
k <- ols_step_backward_aic(model)
k
```

### Plot

```{r stepaicb2, fig.width=5, fig.height=5, fig.align='center'}
model <- lm(y ~ ., data = surgical)
k <- ols_step_backward_aic(model)
plot(k)
```

### Detailed Output

```{r stepwiseaicbdetails}
# stepwise aic backward regression
model <- lm(y ~ ., data = surgical)
ols_step_backward_aic(model, details = TRUE)
```

## Stepwise AIC Regression

Build regression model from a set of candidate predictor variables by entering and removing predictors based on 
Akaike Information Criteria, in a stepwise manner until there is no variable left to enter or remove any more.
The model should include all the candidate predictor variables. If details is set to `TRUE`, each step is displayed.

### Variable Selection

```{r stepwiseaic1}
# stepwise aic regression
model <- lm(y ~ ., data = surgical)
ols_step_both_aic(model)
```

### Plot

```{r stepwiseaic2, fig.width=5, fig.height=5, fig.align='center'}
model <- lm(y ~ ., data = surgical)
k <- ols_step_both_aic(model)
plot(k)
```

### Detailed Output

```{r stepwiseaicdetails}
# stepwise aic regression
model <- lm(y ~ ., data = surgical)
ols_step_both_aic(model, details = TRUE)
```