0

I have an aggregate data frame like below where I know how many customers purchased different cars given the characteristics of each car. How can I create an aggregated logit model (or nested logit model) using this data set? I know about the mlogit and mnlogit package. But can I use them for this purpose?

If not then is there any other way to model logit and nested logit model in R using the following data?

library(data.table)
data("mtcars")
data <- data.table(mtcars)
data[,custo_no:= sample(1:200,nrow(data))]

head(data)
    mpg cyl disp  hp drat    wt  qsec vs am gear carb custo_no
1: 21.0   6  160 110 3.90 2.620 16.46  0  1    4    4      135
2: 21.0   6  160 110 3.90 2.875 17.02  0  1    4    4       24
3: 22.8   4  108  93 3.85 2.320 18.61  1  1    4    1      175
4: 21.4   6  258 110 3.08 3.215 19.44  1  0    3    1       90
5: 18.7   8  360 175 3.15 3.440 17.02  0  0    3    2      180
6: 18.1   6  225 105 2.76 3.460 20.22  1  0    3    1       55
deepAgrawal
  • 673
  • 1
  • 7
  • 25
  • 1
    You threw away the identifying information about the cars when you coerced it to a data.table. – IRTFM Feb 18 '18 at 21:11
  • My actual data is similar to this. I just used mtcars example to give a reproducible example. – deepAgrawal Feb 18 '18 at 21:19
  • The place to start would be the vignettes from the `mlogit`-package. – IRTFM Feb 18 '18 at 21:45
  • I have tried looking at that. But as far as I understand that works for discrete choice models. Where each customer faces multiple choices and chooses one of them. But my problem is an aggregate problem where multiple people choose 1 choice. I don't differentiate between customers. – deepAgrawal Feb 18 '18 at 21:50
  • Discrete choice models is merely the first situation discussed in the vignette. – IRTFM Feb 19 '18 at 01:03

0 Answers0