Error during RRM estimation

[silvia_mauad]

Hello,
I am having problems to estimate my RRM model.
When I run the code bellow, I have this error:

Testing likelihood function...
Error in apollo_preprocess(inputs = mnl_settings, modelType, functionality, :
The "alternatives" argument for model component "model" needs to be a named vector

Can anyone help, please?
Thanks a lot,
Sílvia

My code:

# ################################################################# #
#### LOAD LIBRARY AND DEFINE CORE SETTINGS ####
# ################################################################# #

### Clear memory
#rm(list = ls())


### Load Apollo library

library(apollo)

### Initialise code
apollo_initialise()

### Set core controls
apollo_control = list(
modelName ="mnl_rrm 1",
modelDescr ="Simple MNL model with 30 alternatives",
indivID ="num",
nCores = 6
)


# ################################################################# #
#### LOAD DATA AND APPLY ANY TRANSFORMATIONS ####
# ################################################################# #

### For this example, we generate a database with 2000 observations
# following a logit generation process.
# No panel data (i.e. each obs is from a different individual).
# 100 alternatives, each with two attributes: x1 and x2
# True coefficients are 1 and -1 for x1 and x2, respectively.

N = 4183 #Obs
J = 338 #Alternativas
I = 338
#database <- read.csv("database_com_30_corrigido.csv")

# ################################################################# #
#### DEFINE MODEL PARAMETERS ####
# ################################################################# #

### Vector of parameters, including any that are kept fixed in estimation
apollo_beta=c(b1 = 0, b2 = 0, b3 = 0)

### Vector with names (in quotes) of parameters to be kept fixed at their starting value in apollo_beta, use apollo_beta_fixed = c() if none
apollo_fixed = c()

# ################################################################# #
#### GROUP AND VALIDATE INPUTS ####
# ################################################################# #

apollo_inputs = apollo_validateInputs()

# ################################################################# #
#### DEFINE MODEL AND LIKELIHOOD FUNCTION ####
# ################################################################# #

apollo_probabilities=function(apollo_beta, apollo_inputs, functionality="estimate"){

### Attach inputs and detach after function exit
apollo_attach(apollo_beta, apollo_inputs)
on.exit(apollo_detach(apollo_beta, apollo_inputs))

### Create list of probabilities P
P = list()

### List of utilities: these must use the same names as in mnl_settings, order is irrelevant

V = list()

for(j in 1:J) {

V[[paste0("alt",j)]] =

for(i in 1:I) {

log(2) + log(2) + log(2) +
- get(paste0("AVAIL_",i)) * log(1+exp(b1*(get(paste0("TTi",i))- get(paste0("TTi",j))))) +
- get(paste0("AVAIL_",i)) * log(1+exp(b1*(get(paste0("COi",i))- get(paste0("COi",j))))) +
- get(paste0("AVAIL_",i)) * log(1+exp(b1*(get(paste0("EMPi",i))- get(paste0("EMPi",j)))))
}
}

### Define settings for MNL model component
mnl_settings = list(
alternatives = setNames(1:J, names(V)),
avail = setNames(apollo_inputs$database[,paste0("AVAIL_",1:J)], names(V)),
choiceVar = choice,
V = V
)



### Compute probabilities using MNL model
P[['model']] = apollo_mnl(mnl_settings, functionality)

### Prepare and return outputs of function
P = apollo_prepareProb(P, apollo_inputs, functionality)
return(P)
}

# ################################################################# #
#### MODEL ESTIMATION ####
# ################################################################# #

model_rrm = apollo_estimate(apollo_beta, apollo_fixed, apollo_probabilities, apollo_inputs)

[dpalma]

Hi Sílvia,

I can't say for sure without looking at your data, but I believe there is a problem with the definition of the utilities (V).

I believe the loop should look something like the following:

Code: Select all

for(j in 1:J) {
  V[[paste0("alt",j)]] = 0
    for(i in 1:J) {
      V[[paste0("alt",j)]] = V[[paste0("alt",j)]] - 
        get(paste0("AVAIL_",i)) * log( 1 + exp(b1*get(paste0("TTi" ,i)) - b1*get(paste0("TTi" ,j))) ) -
        get(paste0("AVAIL_",i)) * log( 1 + exp(b2*get(paste0("COi" ,i)) - b2*get(paste0("COi" ,j))) ) -
        get(paste0("AVAIL_",i)) * log( 1 + exp(b3*get(paste0("EMPi",i)) - b3*get(paste0("EMPi",j))) )
    }
}

Please try that. If that doesn't work, I would need to look at least to some of the data to understand what else could be going wrong. If the data is sensitive, you could email the headers and a handful of rows to me to: D.Palma [at] leeds.ac.uk

Cheers
David