Important: Read this before posting to this forum

  1. This forum is for questions related to the use of Apollo. We will answer some general choice modelling questions too, where appropriate, and time permitting. We cannot answer questions about how to estimate choice models with other software packages.
  2. There is a very detailed manual for Apollo available at http://www.ApolloChoiceModelling.com/manual.html. This contains detailed descriptions of the various Apollo functions, and numerous examples are available at http://www.ApolloChoiceModelling.com/examples.html. In addition, help files are available for all functions, using e.g. ?apollo_mnl
  3. Before asking a question on the forum, users are kindly requested to follow these steps:
    1. Check that the same issue has not already been addressed in the forum - there is a search tool.
    2. Ensure that the correct syntax has been used. For any function, detailed instructions are available directly in Apollo, e.g. by using ?apollo_mnl for apollo_mnl
    3. Check the frequently asked questions section on the Apollo website, which discusses some common issues/failures. Please see http://www.apollochoicemodelling.com/faq.html
    4. Make sure that R is using the latest official release of Apollo.
  4. If the above steps do not resolve the issue, then users should follow these steps when posting a question:
    1. provide full details on the issue, including the entire code and output, including any error messages
    2. posts will not immediately appear on the forum, but will be checked by a moderator first. This may take a day or two at busy times. There is no need to submit the post multiple times.

Error in estimating Latent class model with RP-SP data

Ask questions about errors you encouunter. Please make sure to include full details about your model specifications, and ideally your model file.
Post Reply
Aditya249
Posts: 16
Joined: 27 Jan 2023, 08:47

Error in estimating Latent class model with RP-SP data

Post by Aditya249 »

Hi!
Prof. Hess, and Dr. Palma

I am trying to develop a latent class model with Joint RP-SP data. However, I am getting the following error: object 'mnl_settings_RP' not found

Could you please help me out in resolving this?

The code is as follows:

Code: Select all

# ################################################################# #
#### 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       = "Apollo_IPA",
  modelDescr      = "RP_SP_HCM",
  indivID         = "ID", 
  nCores          = 16,
  outputDirectory = "output"
)

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

database = read.csv("RP_SP_CP_3.csv")

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

### Vector of parameters, including any that are kept fixed in estimation
apollo_beta=c(asc_local               = 0,
              asc_metro               = 0,
              asc_car                 = 0,
              asc_tw                  = 0,

              b_tt_a                    = 0,
              b_tc_a                    = 0,
              b_at_a                    = 0,

              b_tt_b                    = 0,
              b_tc_b                    = 0,
              b_at_b                    = 0,

              delta_a                   = 0,
              delta_b                   = 0,

              female_a                  = 0,
              Income_0_50k_a            = 0,
              Age_18_30_a               = 0,

              female_b                  = 0,
              Income_0_50k_b            = 0,
              Age_18_30_b               = 0,
            

              mu_RP                   = 1,
              mu_SP                   = 1)

### 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("mu_RP", "asc_car", "female_a")


# ################################################################# #
#### DEFINE LATENT CLASS COMPONENTS                              ####
# ################################################################# #

apollo_lcPars=function(apollo_beta, apollo_inputs){
  lcpars = list()
  lcpars[["b_tt"]] = list(b_tt_a, b_tt_b)
  lcpars[["b_tc"]] = list(b_tc_a, b_tc_b)
  lcpars[["b_at"]] = list(b_at_a, b_at_b)

### Utilities of class allocation model
  V=list()


  V[["class_a"]] = delta_a + female_a*female + Income_0_50k_a*Income_0_50k + Age_18_30_a*Age_18_30
  V[["class_b"]] = delta_b + female_b*female + Income_0_50k_b*Income_0_50k + Age_18_30_b*Age_18_30

### Settings for class allocation models
  classAlloc_settings = list(
    classes      = c(class_a=1, class_b=2), 
    utilities    = V  
  )
  
  lcpars[["pi_values"]] = apollo_classAlloc(classAlloc_settings)
  
  return(lcpars)
}

# ################################################################# #
#### 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()

  ### Loop over classes
  for(s in 1:2){

 ### List of utilities (before applying scales): these must use the same names as in mnl_settings, order is irrelevant
  V = list()
  V[["local"]]  = asc_local + b_tt[[s]]*tt_local + b_tc[[s]]*tc_local + b_at[[s]]*at_local


  V[["metro"]]  = asc_metro  + b_tt[[s]]*tt_metro + b_tc[[s]]*tc_metro + b_at[[s]]*at_metro


  V[["car"]]    = asc_car    + b_tt[[s]]*tt_car + b_tc[[s]]*tc_car        
                  

  V[["tw"]]     = asc_tw     + b_tt[[s]]*tt_tw + b_tc[[s]]*tc_tw 


    mnl_settings_RP$componentName = paste0("Class_",s)
    mnl_settings_SP$componentName = paste0("Class_",s)

### Compute probabilities for the RP part of the data using MNL model
  mnl_settings_RP = list(
       alternatives  = c(local=1, metro=2, car=3, tw=4), 
       avail         = list(local=av_local, metro=av_metro, car=av_car, tw=av_tw), 
       choiceVar     = Choice, 
       utilities     = list(local = mu_RP*V[["local"]],
                            metro = mu_RP*V[["metro"]],
                            car   = mu_RP*V[["car"]],
                            tw    = mu_RP*V[["tw"]]),
       rows          = (RP==1)

  )

    mnl_settings_RP$componentName = paste0("Class_",s)


 P[["RP"]] = apollo_mnl(mnl_settings_RP, functionality)

### Compute probabilities for the RP part of the data using MNL model
  mnl_settings_SP = list(
       alternatives  = c(local=1, metro=2, car=3, tw=4), 
       avail         = list(local=av_local, metro=av_metro, car=av_car, tw=av_tw), 
       choiceVar     = Choice, 
       utilities     = list(local  = mu_SP*V[["local"]],
                            metro  = mu_SP*V[["metro"]],
                            car    = mu_SP*V[["car"]],
                            tw     = mu_SP*V[["tw"]]),
       rows          = (SP==1)

  )

    mnl_settings_SP$componentName = paste0("Class_",s)

 P[["SP"]] = apollo_mnl(mnl_settings_SP, functionality)
   



### Compute within-class choice probabilities using MNL model
    P[[paste0("Class_",s)]] = apollo_mnl(mnl_settings_RP, functionality)
    P[[paste0("Class_",s)]] = apollo_mnl(mnl_settings_SP, functionality)
    
    ### Take product across observation for same individual
    P[[paste0("Class_",s)]] = apollo_panelProd(P[[paste0("Class_",s)]], apollo_inputs ,functionality)
    
  }

  ### Combined model
  P = apollo_combineModels(P, apollo_inputs, functionality)

 ### Compute latent class model probabilities
  lc_settings  = list(inClassProb = P, classProb=pi_values)
  P[["RP"]] = apollo_lc(lc_settings, apollo_inputs, functionality)
  P[["SP"]] = apollo_lc(lc_settings, apollo_inputs, functionality)

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


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

### Estimate model
model = apollo_estimate(apollo_beta, apollo_fixed, 
                        apollo_probabilities, apollo_inputs)

### Show output in screen
apollo_modelOutput(model)

### Save output to file(s)


# ################################################################# #
#### MODEL OUTPUTS                                               ####
# ################################################################# #

# ----------------------------------------------------------------- #
#---- FORMATTED OUTPUT (TO SCREEN)                               ----
# ----------------------------------------------------------------- #

apollo_modelOutput(model)
Top
stephanehess
Site Admin
Posts: 1049
Joined: 24 Apr 2020, 16:29

Re: Error in estimating Latent class model with RP-SP data

Post by stephanehess »

Hi

simple fix. Have a look at your code. You have

Code: Select all

mnl_settings_RP$componentName = paste0("Class_",s)
mnl_settings_SP$componentName = paste0("Class_",s)
before you have:

Code: Select all

mnl_settings_RP = list(
       alternatives  = c(local=1, metro=2, car=3, tw=4), 
...
so the list mnl_settings_RP does not yet exist when you try to set the componentName element inside it

Stephane
--------------------------------
Stephane Hess
www.stephanehess.me.uk
Top
Aditya249
Posts: 16
Joined: 27 Jan 2023, 08:47

Re: Error in estimating Latent class model with RP-SP data

Post by Aditya249 »

Hi!
Thank you for your prompt response. Based on your suggestion, I have revised the code. However, now I am getting the following error:

Error in apollo_combineModels(P, apollo_inputs, functionality) :
SPECIFICATION ISSUE - A component called model already exists in P before calling apollo_combineModels!

Please help in resolving this.

The revised code is as follows:

Code: Select all

# ################################################################# #
#### 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       = "Apollo_IPA",
  modelDescr      = "RP_SP_HCM",
  indivID         = "ID", 
  nCores          = 16,
  outputDirectory = "output"
)

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

database = read.csv("RP_SP_CP_3.csv")

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

### Vector of parameters, including any that are kept fixed in estimation
apollo_beta=c(asc_local               = 0,
              asc_metro               = 0,
              asc_car                 = 0,
              asc_tw                  = 0,

              b_tt_a                    = 0,
              b_tc_a                    = 0,
              b_at_a                    = 0,

              b_tt_b                    = 0,
              b_tc_b                    = 0,
              b_at_b                    = 0,

              delta_a                   = 0,
              delta_b                   = 0,

              female_a                  = 0,
              Income_0_50k_a            = 0,
              Age_18_30_a               = 0,

              female_b                  = 0,
              Income_0_50k_b            = 0,
              Age_18_30_b               = 0,
            

              mu_RP                   = 1,
              mu_SP                   = 1)

### 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("mu_RP", "asc_car", "female_a")


# ################################################################# #
#### DEFINE LATENT CLASS COMPONENTS                              ####
# ################################################################# #

apollo_lcPars=function(apollo_beta, apollo_inputs){
  lcpars = list()
  lcpars[["b_tt"]] = list(b_tt_a, b_tt_b)
  lcpars[["b_tc"]] = list(b_tc_a, b_tc_b)
  lcpars[["b_at"]] = list(b_at_a, b_at_b)

### Utilities of class allocation model
  V=list()


  V[["class_a"]] = delta_a + female_a*female + Income_0_50k_a*Income_0_50k + Age_18_30_a*Age_18_30
  V[["class_b"]] = delta_b + female_b*female + Income_0_50k_b*Income_0_50k + Age_18_30_b*Age_18_30

### Settings for class allocation models
  classAlloc_settings = list(
    classes      = c(class_a=1, class_b=2), 
    utilities    = V  
  )
  
  lcpars[["pi_values"]] = apollo_classAlloc(classAlloc_settings)
  
  return(lcpars)
}

# ################################################################# #
#### 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()

  ### Loop over classes
  for(s in 1:2){

 ### List of utilities (before applying scales): these must use the same names as in mnl_settings, order is irrelevant
  V = list()
  V[["local"]]  = asc_local + b_tt[[s]]*tt_local + b_tc[[s]]*tc_local + b_at[[s]]*at_local


  V[["metro"]]  = asc_metro  + b_tt[[s]]*tt_metro + b_tc[[s]]*tc_metro + b_at[[s]]*at_metro


  V[["car"]]    = asc_car    + b_tt[[s]]*tt_car + b_tc[[s]]*tc_car        
                  

  V[["tw"]]     = asc_tw     + b_tt[[s]]*tt_tw + b_tc[[s]]*tc_tw 


### Compute probabilities for the RP part of the data using MNL model
  mnl_settings_RP = list(
       alternatives  = c(local=1, metro=2, car=3, tw=4), 
       avail         = list(local=av_local, metro=av_metro, car=av_car, tw=av_tw), 
       choiceVar     = Choice, 
       utilities     = list(local = mu_RP*V[["local"]],
                            metro = mu_RP*V[["metro"]],
                            car   = mu_RP*V[["car"]],
                            tw    = mu_RP*V[["tw"]]),
       rows          = (RP==1)

  )


 P[["RP"]] = apollo_mnl(mnl_settings_RP, functionality)

   

### Compute probabilities for the RP part of the data using MNL model
  mnl_settings_SP = list(
       alternatives  = c(local=1, metro=2, car=3, tw=4), 
       avail         = list(local=av_local, metro=av_metro, car=av_car, tw=av_tw), 
       choiceVar     = Choice, 
       utilities     = list(local  = mu_SP*V[["local"]],
                            metro  = mu_SP*V[["metro"]],
                            car    = mu_SP*V[["car"]],
                            tw     = mu_SP*V[["tw"]]),
       rows          = (SP==1)

  )

 P[["SP"]] = apollo_mnl(mnl_settings_SP, functionality)



mnl_settings_RP$componentName = paste0("Class_",s)    
mnl_settings_SP$componentName = paste0("Class_",s)

   
### Compute within-class choice probabilities using MNL model
    P[[paste0("Class_",s)]] = apollo_mnl(mnl_settings_RP, functionality)
    P[[paste0("Class_",s)]] = apollo_mnl(mnl_settings_SP, functionality)

 
### Combined model
    P = apollo_combineModels(P, apollo_inputs, functionality)


### Take product across observation for same individual
    P[[paste0("Class_",s)]] = apollo_panelProd(P[[paste0("Class_",s)]], apollo_inputs ,functionality)

    P = apollo_panelProd(P, apollo_inputs, functionality)

}
    
 ### Compute latent class model probabilities
  lc_settings  = list(inClassProb = P, classProb=pi_values)
  P[["model"]] = apollo_lc(lc_settings, apollo_inputs, functionality)


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


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

### Estimate model
model = apollo_estimate(apollo_beta, apollo_fixed, 
                        apollo_probabilities, apollo_inputs)

### Show output in screen
apollo_modelOutput(model)

### Save output to file(s)


# ################################################################# #
#### MODEL OUTPUTS                                               ####
# ################################################################# #

# ----------------------------------------------------------------- #
#---- FORMATTED OUTPUT (TO SCREEN)                               ----
# ----------------------------------------------------------------- #

apollo_modelOutput(model)
Top
stephanehess
Site Admin
Posts: 1049
Joined: 24 Apr 2020, 16:29

Re: Error in estimating Latent class model with RP-SP data

Post by stephanehess »

You'll need to be a bit careful here as combineModels is used inside the LC. probably best to work with temporary lists rather than P for RP and SP. You also need to tell it which ones to combine

Something like this (look where I have used P_temp and where I have used P

You probably need to make some more changes to this, but it's a start for you

Code: Select all

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()
[color=#FF0000]  P_temp = list()[/color]

  ### Loop over classes
  for(s in 1:2){

 ### List of utilities (before applying scales): these must use the same names as in mnl_settings, order is irrelevant
  V = list()
  V[["local"]]  = asc_local + b_tt[[s]]*tt_local + b_tc[[s]]*tc_local + b_at[[s]]*at_local
  V[["metro"]]  = asc_metro  + b_tt[[s]]*tt_metro + b_tc[[s]]*tc_metro + b_at[[s]]*at_metro
  V[["car"]]    = asc_car    + b_tt[[s]]*tt_car + b_tc[[s]]*tc_car        
  V[["tw"]]     = asc_tw     + b_tt[[s]]*tt_tw + b_tc[[s]]*tc_tw 


### Compute probabilities for the RP part of the data using MNL model
  mnl_settings_RP = list(
       alternatives  = c(local=1, metro=2, car=3, tw=4), 
       avail         = list(local=av_local, metro=av_metro, car=av_car, tw=av_tw), 
       choiceVar     = Choice, 
       utilities     = list(local = mu_RP*V[["local"]],
                            metro = mu_RP*V[["metro"]],
                            car   = mu_RP*V[["car"]],
                            tw    = mu_RP*V[["tw"]]),
       rows          = (RP==1)

  )

 [color=#FF0000]P_temp[[paste0("RP_class_",s)]][/color] = apollo_mnl(mnl_settings_RP, functionality)

### Compute probabilities for the RP part of the data using MNL model
  mnl_settings_SP = list(
       alternatives  = c(local=1, metro=2, car=3, tw=4), 
       avail         = list(local=av_local, metro=av_metro, car=av_car, tw=av_tw), 
       choiceVar     = Choice, 
       utilities     = list(local  = mu_SP*V[["local"]],
                            metro  = mu_SP*V[["metro"]],
                            car    = mu_SP*V[["car"]],
                            tw     = mu_SP*V[["tw"]]),
       rows          = (SP==1)

  )

 [color=#FF0000]P_temp[[paste0("SP_class_",s)]][/color] = apollo_mnl(mnl_settings_SP, functionality)

### Combined model
[color=#FF0000]    PP[[paste0("Class_",s)]] = apollo_combineModels(P_temp, apollo_inputs, functionality,components=c(paste0("RP_class_",s),paste0("SP_class_",s)))[/color]

### Take product across observation for same individual
[color=#FF0000]    P[[paste0("Class_",s)]] = apollo_panelProd(P[[paste0("Class_",s)]], apollo_inputs ,functionality)[/color]

[color=#FF0000]    P[[paste0("Class_",s)]] = apollo_panelProd(P[[paste0("Class_",s)]], apollo_inputs, functionality)[/color]

}
    
 ### Compute latent class model probabilities
  lc_settings  = list(inClassProb = P, classProb=pi_values)
  P[["model"]] = apollo_lc(lc_settings, apollo_inputs, functionality)


  ### Prepare and return outputs of function
  P = apollo_prepareProb(P, apollo_inputs, functionality)
  return(P)
}
--------------------------------
Stephane Hess
www.stephanehess.me.uk
Top
Aditya249
Posts: 16
Joined: 27 Jan 2023, 08:47

Re: Error in estimating Latent class model with RP-SP data

Post by Aditya249 »

Hi!

I tried multiple times to work on this code and modify it based on the provided suggestion. Now I am getting the following error: Testing influence of parametersError in apollo_estimate(apollo_beta, apollo_fixed, apollo_probabilities, :
SPECIFICATION ISSUE - Parameters b_tt_a, b_tc_a, b_at_a, female_b do not influence the log-likelihood of your model!

I request you to please help in resolving this.

The modified code is as follows:
# ################################################################# #
#### 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 = "Apollo_IPA",
modelDescr = "RP_SP_HCM",
indivID = "ID",
nCores = 16,
outputDirectory = "output"
)

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

database = read.csv("RP_SP_CP_3.csv")

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

### Vector of parameters, including any that are kept fixed in estimation
apollo_beta=c(asc_local = 0,
asc_metro = 0,
asc_car = 0,
asc_tw = 0,

b_tt_a = 0,
b_tc_a = 0,
b_at_a = 0,

b_tt_b = 0,
b_tc_b = 0,
b_at_b = 0,

delta_a = 0,
delta_b = 0,

female_a = 0,
Income_0_50k_a = 0,
Age_18_30_a = 0,

female_b = 0,
Income_0_50k_b = 0,
Age_18_30_b = 0,


mu_RP = 1,
mu_SP = 1)

### 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("mu_RP", "asc_car", "female_a")


# ################################################################# #
#### DEFINE LATENT CLASS COMPONENTS ####
# ################################################################# #

apollo_lcPars=function(apollo_beta, apollo_inputs){
lcpars = list()
lcpars[["b_tt"]] = list(b_tt_a, b_tt_b)
lcpars[["b_tc"]] = list(b_tc_a, b_tc_b)
lcpars[["b_at"]] = list(b_at_a, b_at_b)

### Utilities of class allocation model
V=list()
V[["class_a"]] = delta_a + female_a*female + Income_0_50k_a*Income_0_50k + Age_18_30_a*Age_18_30
V[["class_b"]] = delta_b + female_b*female + Income_0_50k_b*Income_0_50k + Age_18_30_b*Age_18_30

### Settings for class allocation models
classAlloc_settings = list(
classes = c(class_a=1, class_b=2),
utilities = V
)

lcpars[["pi_values"]] = apollo_classAlloc(classAlloc_settings)

return(lcpars)
}

# ################################################################# #
#### 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()
P_temp = list()


### Loop over classes
for(s in 1:2){

### List of utilities (before applying scales): these must use the same names as in mnl_settings, order is irrelevant
V = list()
V[["local"]] = asc_local + b_tt[[s]]*tt_local + b_tc[[s]]*tc_local + b_at[[s]]*at_local
V[["metro"]] = asc_metro + b_tt[[s]]*tt_metro + b_tc[[s]]*tc_metro + b_at[[s]]*at_metro
V[["car"]] = asc_car + b_tt[[s]]*tt_car + b_tc[[s]]*tc_car
V[["tw"]] = asc_tw + b_tt[[s]]*tt_tw + b_tc[[s]]*tc_tw


### Compute probabilities for the RP part of the data using MNL model
mnl_settings_RP = list(
alternatives = c(local=1, metro=2, car=3, tw=4),
avail = list(local=av_local, metro=av_metro, car=av_car, tw=av_tw),
choiceVar = Choice,
utilities = list(local = mu_RP*V[["local"]],
metro = mu_RP*V[["metro"]],
car = mu_RP*V[["car"]],
tw = mu_RP*V[["tw"]]),
rows = (RP==1)

)

P_temp[[paste0("RP_class_",s)]] = apollo_mnl(mnl_settings_RP, functionality)

### Compute probabilities for the RP part of the data using MNL model
mnl_settings_SP = list(
alternatives = c(local=1, metro=2, car=3, tw=4),
avail = list(local=av_local, metro=av_metro, car=av_car, tw=av_tw),
choiceVar = Choice,
utilities = list(local = mu_SP*V[["local"]],
metro = mu_SP*V[["metro"]],
car = mu_SP*V[["car"]],
tw = mu_SP*V[["tw"]]),
rows = (SP==1)

)

P_temp[[paste0("SP_class_",s)]] = apollo_mnl(mnl_settings_SP, functionality)


### Combined model
P[[paste0("Class_",s)]] = apollo_combineModels(P_temp, apollo_inputs, functionality,components=c(paste0("RP_class_",s),paste0("SP_class_",s)))

### Take product across observation for same individual
P[[paste0("Class_",s)]] = apollo_panelProd(P[[paste0("Class_",s)]], apollo_inputs ,functionality)


}

### Compute latent class model probabilities
lc_settings = list(inClassProb = P, classProb=pi_values)
P[["model"]] = apollo_lc(lc_settings, apollo_inputs, functionality)


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


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

### Estimate model
model = apollo_estimate(apollo_beta, apollo_fixed,
apollo_probabilities, apollo_inputs)

### Show output in screen
apollo_modelOutput(model)

### Save output to file(s)


# ################################################################# #
#### MODEL OUTPUTS ####
# ################################################################# #

# ----------------------------------------------------------------- #
#---- FORMATTED OUTPUT (TO SCREEN) ----
# ----------------------------------------------------------------- #

apollo_modelOutput(model)
Top
stephanehess
Site Admin
Posts: 1049
Joined: 24 Apr 2020, 16:29

Re: Error in estimating Latent class model with RP-SP data

Post by stephanehess »

Can you share your code and data with me outside the forum, and I'll try to debug it for you
--------------------------------
Stephane Hess
www.stephanehess.me.uk
Top
Aditya249
Posts: 16
Joined: 27 Jan 2023, 08:47

Re: Error in estimating Latent class model with RP-SP data

Post by Aditya249 »

Thank You for offering the help, Prof. Hess.

As suggested, I have shared the code and data with you at stephane.hess@gmail.com
Top
stephanehess
Site Admin
Posts: 1049
Joined: 24 Apr 2020, 16:29

Re: Error in estimating Latent class model with RP-SP data

Post by stephanehess »

Hi

two changes are needed in your code.

As apollo_combineModels is called inside the latent class model, you need

Code: Select all

P[[paste0("Class_",s)]] = apollo_combineModels(P_temp, apollo_inputs, functionality,components=c(paste0("RP_class_",s),paste0("SP_class_",s)),asList=FALSE)
instead of

Code: Select all

P[[paste0("Class_",s)]] = apollo_combineModels(P_temp, apollo_inputs, functionality,components=c(paste0("RP_class_",s),paste0("SP_class_",s)))
Also, your model is currently overspecified. You cannot estimate the parameters in the class allocation model for both classes. So you need

Code: Select all

apollo_fixed = c("mu_RP", "asc_car", 
                 "delta_a",
                 "female_a",      
                 "Income_0_50k_a",
                 "Age_18_30_a"   )
Stephane
--------------------------------
Stephane Hess
www.stephanehess.me.uk
Top
Aditya249
Posts: 16
Joined: 27 Jan 2023, 08:47

Re: Error in estimating Latent class model with RP-SP data

Post by Aditya249 »

Thank you Prof. Hess for the suggestion.

The code is working fine after altering the starting values of those parameters.
Top
stephanehess
Site Admin
Posts: 1049
Joined: 24 Apr 2020, 16:29

Re: Error in estimating Latent class model with RP-SP data

Post by stephanehess »

yes, having all the starting values at zero also led to an issue with the initial check of the gradients. It's always good practice to use different starting values in different classes
--------------------------------
Stephane Hess
www.stephanehess.me.uk
Top
Post Reply

Return to “Errors during model validation and/or estimation.”