R 中的通用逻辑回归

Question

我想知道您是否可以创建逻辑回归的通用版本，我的意思是允许在预测数据集中使用不同名称的变量。

Answer 1

require(mtcars) # this represents your training data
train <- mtcars

logit <- glm(vs~mpg, data = train, family = "binomial")

generic.logit <- function(mpg, data=test){
  test          <- data
  test$mpg      <- mpg
  mypredictions <- predict(logit, test)
  return(mypredictions)
}

test <- mtcars
test$random_name <- test$mpg
test$mpg         <- NULL # this represents your prediction data set

generic.logit(data=test, mpg = test$random_name)



      Mazda RX4       Mazda RX4 Wag          Datsun 710      Hornet 4 Drive   Hornet Sportabout             Valiant 
      0.2056113           0.2056113           0.9803557           0.3777768          -0.7843397          -1.0425879 
     Duster 360           Merc 240D            Merc 230            Merc 280           Merc 280C          Merc 450SE 
     -2.6781592           1.6690173           0.9803557          -0.5691330          -1.1717119          -1.7742908 
     Merc 450SL         Merc 450SLC  Cadillac Fleetwood Lincoln Continental   Chrysler Imperial            Fiat 128 
     -1.3869187          -2.2907871          -4.3567720          -4.3567720          -2.5059938           5.1123255 
    Honda Civic      Toyota Corolla       Toyota Corona    Dodge Challenger         AMC Javelin          Camaro Z28 
      4.2514984           5.7579458           0.4208181          -2.1616630          -2.2907871          -3.1085728 
   Pontiac Firebird           Fiat X1-9       Porsche 914-2        Lotus Europa      Ford Pantera L        Ferrari Dino 
         -0.5691330           2.9172165           2.3576790           4.2514984          -2.0325390          -0.3539262 
      Maserati Bora          Volvo 142E 
         -2.3768698           0.3777768

R 中的通用逻辑回归

Generic Logistic Regression in R

r

logistic-regression