Inicializa

library (ISLR)

str(Auto)

## 'data.frame':    392 obs. of  9 variables:
##  $ mpg         : num  18 15 18 16 17 15 14 14 14 15 ...
##  $ cylinders   : num  8 8 8 8 8 8 8 8 8 8 ...
##  $ displacement: num  307 350 318 304 302 429 454 440 455 390 ...
##  $ horsepower  : num  130 165 150 150 140 198 220 215 225 190 ...
##  $ weight      : num  3504 3693 3436 3433 3449 ...
##  $ acceleration: num  12 11.5 11 12 10.5 10 9 8.5 10 8.5 ...
##  $ year        : num  70 70 70 70 70 70 70 70 70 70 ...
##  $ origin      : num  1 1 1 1 1 1 1 1 1 1 ...
##  $ name        : Factor w/ 304 levels "amc ambassador brougham",..: 49 36 231 14 161 141 54 223 241 2 ...

variables <-
  data.frame(dependiente = names(Auto)[1],
             independientes = names(Auto)[2:(ncol(Auto) - 1)])

set.seed (1)
train=sample (nrow(Auto) ,nrow(Auto)*0.7)

Función

stepwise_forward <- function(dependiente,independientes,dependiente_f){
  
  if(dependiente_f == 0){
    dependiente_f <- paste(dependiente,"~", sep = "")
  }
  r2 <- c(r2=rep(1:length(independientes)))
  ar2 <- c(r2=rep(1:length(independientes)))
  rss <- c(rss=rep(1:length(independientes)))
  r22 <- 0
  formu <- ""
  res <- data.frame(independientes,r2,rss,formu)[1,]
  
  for (i in 1:length(independientes)) {
    temp <- paste(dependiente_f,independientes[i],collapse = "+")
    # print(paste("formula:",temp))
    
    temp.fit<-lm(temp, data=Auto, subset = train)
    r2[i] <- summary(temp.fit)$r.squared
    rss[i] <- deviance(temp.fit)
    ar2[i] <- summary(temp.fit)$adj.r.squared
    
    if(r22 == 0){
      r22 <- summary(temp.fit)$r.squared
      formu <- temp
      # print(paste("menor r2:", r22))
    }else {
      if ( summary(temp.fit)$r.squared > r22) {
        r22 <- summary(temp.fit)$r.squared
        formu <- temp
        # print(paste("menor r2:", r22))
      } else {
        0
      }
    }
  }  
  
  temp <- data.frame(independientes,r2,rss,ar2,formu)
  
  temp
  
  print(paste("selecciona:", temp[temp$r2 == r22,]$independientes))
  print(paste("formula:",formu))
  res <- temp[temp$r2 == r22,]
  
  selec <- independientes==temp[temp$r2 == r22,]$independientes
  independientes <- independientes[!selec]
  
  if(is.na(independientes[1]) ){
    dependiente_f <- paste(dependiente_f,temp[temp$r2 == r22,]$independientes, sep = "")
    return(res)
  } else {
    dependiente_f <- paste(dependiente_f,temp[temp$r2 == r22,]$independientes,"+", sep = "")
    res <- rbind(res,stepwise_forward(dependiente,independientes,dependiente_f))
  }
}

Llamada a la funcion

modelos <- stepwise_forward(unique(variables$dependiente),variables$independientes,0)

## [1] "selecciona: weight"
## [1] "formula: mpg~ weight"
## [1] "selecciona: year"
## [1] "formula: mpg~weight+ year"
## [1] "selecciona: origin"
## [1] "formula: mpg~weight+year+ origin"
## [1] "selecciona: horsepower"
## [1] "formula: mpg~weight+year+origin+ horsepower"
## [1] "selecciona: displacement"
## [1] "formula: mpg~weight+year+origin+horsepower+ displacement"
## [1] "selecciona: cylinders"
## [1] "formula: mpg~weight+year+origin+horsepower+displacement+ cylinders"
## [1] "selecciona: acceleration"
## [1] "formula: mpg~weight+year+origin+horsepower+displacement+cylinders+ acceleration"

Resultados

str(modelos)

## 'data.frame':    7 obs. of  5 variables:
##  $ independientes: Factor w/ 7 levels "acceleration",..: 6 7 5 4 3 2 1
##  $ r2            : num  0.691 0.806 0.815 0.817 0.818 ...
##  $ rss           : num  5099 3201 3047 3022 3007 ...
##  $ ar2           : num  0.69 0.804 0.813 0.814 0.814 ...
##  $ formu         : Factor w/ 7 levels "mpg~ weight",..: 1 2 3 4 5 6 7

modelos

##      independientes        r2      rss       ar2
## r24          weight 0.6906605 5099.450 0.6895232
## r25            year 0.8058003 3201.375 0.8043671
## r251         origin 0.8151730 3046.866 0.8131194
## r23      horsepower 0.8167047 3021.616 0.8139791
## r22    displacement 0.8175892 3007.036 0.8141860
## r21       cylinders 0.8194478 2976.396 0.8153905
## r2     acceleration 0.8194568 2976.248 0.8147057
##                                                                       formu
## r24                                                             mpg~ weight
## r25                                                        mpg~weight+ year
## r251                                                mpg~weight+year+ origin
## r23                                      mpg~weight+year+origin+ horsepower
## r22                         mpg~weight+year+origin+horsepower+ displacement
## r21               mpg~weight+year+origin+horsepower+displacement+ cylinders
## r2   mpg~weight+year+origin+horsepower+displacement+cylinders+ acceleration