Predikce podílů pomocí hierarchických modelů

Modely výběru pro marketing v R

Elea McDonnell Feit

Assistant Professor of Marketing, Drexel University

Hierarchický model s korelacemi

m10 <- mlogit(choice ~ 0 + seat + trans + convert + price, 
              data = sportscar, 
              rpar = myrpar, panel = TRUE, correlation = TRUE)

cor.mlogit(m10)

                 seat2      seat4 transmanual convertyes      price
seat2        1.0000000 -0.0007873  -0.5779923  0.6584917 -0.6665061
seat4       -0.0007873  1.0000000   0.2118747 -0.2057497  0.5457144
transmanual -0.5779923  0.2118747   1.0000000 -0.7203588  0.2832574
convertyes   0.6584917 -0.2057497  -0.7203588  1.0000000 -0.3387832
price       -0.6665061  0.5457144   0.2832574 -0.3387832  1.0000000

Produkty, pro které chceme predikovat podíly

prod

  seat  trans convert price
1    2 manual      no    35
2    2   auto      no    30

prod.coded

  seat2 seat4 transmanual convertyes price
1     1     0           1          0    35
2     1     0           0          0    30

Predikce podílů pro hierarchický model

mean <- m10$coef[1:5]   # Hard coded
Sigma <- cov.mlogit(m10)

share <- matrix(NA, nrow = 1000, ncol = nrow(prod.coded))
for (i in 1:1000) {

  coef <- mvrnorm(1, mu = mean, Sigma = Sigma)

  utility <- prod.coded %*% coef

  share[i,] <- exp(utility) / sum(exp(utility))

}

cbind(colMeans(share), prod)

  colMeans(share) seat  trans convert price
1       0.1279492    2 manual      no    35
2       0.8720508    2   auto      no    30

Pojďme cvičit!

Modely výběru pro marketing v R