對其他特徵偏好的異質性

R 的行銷選擇模型

Elea McDonnell Feit

Assistant Professor of Marketing, Drexel University

因子編碼的另一種方式

虛擬編碼（我們之前用的）

  seat4 seat5
2     0     0
4     1     0
5     0     1

效果編碼（較適合階層式模型）

   seat2  seat4
2      1      0
4      0      1
5     -1     -1

變更因子的編碼

contrasts(sportscar$seat) <- contr.sum(levels(sportscar$seat))

dimnames(contrasts(sportscar$seat))[[2]] <- levels(sportscar$seat)[1:2]

contrasts(sportscar$seat)

讓所有係數皆具異質性

my_rpar <- c("n", "n", "n", "n", "n")

m3 <- mlogit(choice ~ 0 + seat + trans + convert + price,
             data = sportscar)
names(my_rpar) <- names(m3$coefficients)

my_rpar

      seat2       seat4 transmanual  convertyes       price 
        "n"         "n"         "n"         "n"         "n"

m8 <- mlogit(choice ~ 0 + seat + trans + convert + price, 
             data = sportscar, 
             panel = TRUE, rpar = my_rpar)

階層式模型的參數

m8 <- mlogit(choice ~ 0 + seat + trans + convert + price, 
              data = sportscar, 
             panel = TRUE, rpar = my_rpar)
plot(m8, par = c("seat2", "seat4"))

效果編碼參數分佈

基準水準的係數

m8$coef[1:2]

     seat2      seat4 
-0.1686077 -0.1912170

-sum(m8$coef[1:2])

[1] 0.3598248

用 `chocolate` 資料來試試吧！

R 的行銷選擇模型