Heterogeneity in preference for other features

Choice Modeling for Marketing in R

Elea McDonnell Feit

Assistant Professor of Marketing, Drexel University

A different way to code factors

Dummy coding (what we've been doing)

  seat4 seat5
2     0     0
4     1     0
5     0     1

Effects coding (better for hierarchical models)

   seat4  seat5
2     -1     -1
4      1      0
5      0      1

Changing the coding for a factor

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

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

contrasts(sportscar$seat)

Making all the coefficients heterogeneous

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

seat4       seat5 transmanual  convertyes       price 
  "n"         "n"         "n"         "n"         "n"

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

Hierarchical model parameters

m8 <- mlogit(choice ~ 0 + seat + trans + convert + price, 
              data = sportscar, 
             panel = TRUE, rpar = my_rpar)
plot(m8, par = c("seat4", "seat5")

effect coded parameter distributions

Coefficient for the base level

m8$coef[1:2]

     seat4      seat5 
-0.1852167  0.3519204

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

-0.1667037

Let's try it with the `chocolate` data!

Choice Modeling for Marketing in R