Page 1 of 1

Latent Class Choice Model: NaNs produced

Posted: 04 Nov 2025, 10:59
by kunbeto
Dear Dr. Stephane and the Apollo Choice Modelling Community,

I am trying to configure a 3-class LCCM model incorporating a Latent variable and running into an error of convergence to a saddle point (NaNs produced).
Fundamentally, the choice-specific model consists of 5 alternatives (including an opt-out-like alternative, without any specific alternative attributes). No covariates are considered at this stage. I am sharing the code and the output as follows:

1. The code

Code: Select all

library(apollo)

### Initialise code
apollo_initialise()

### Set core controls
apollo_control = list(
  modelName  = "MNL",
  modelDescr = "Simple MNL model MaaS bundle SP x LCCM 3C x LV",
  indivID    = "ID",
  nCores     = 16,
  panelData = TRUE,
  noValidation = FALSE,
  analyticGrad = FALSE,
  outputDirectory = "output"
)

# ####################################################### #
#### 2. Data loading                                   ####
# ####################################################### #
maas <- read.csv("HCM_Data_R_working.csv")
database = maas
rm(maas)

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

### Vector of parameters, including any that are kept fixed 
### during estimation
apollo_beta=c(asc_B1_a    = 1.3, asc_B1_b     = -0.91, asc_B1_c     = 0.35,
              asc_B2_a    = 2.0, asc_B2_b     = -1.65, asc_B2_c    = 2.0,
              asc_B3_a    = 1.9, asc_B3_b     = 1.17, asc_B3_c     = 1.89,
              asc_B4_a    = 1.06, asc_B4_b   = 2.15, asc_B4_c     = 1.61,
              asc_PAYG    = 0,
              # sigma_eta   = 1,
              # Class allocation parameters
              delta_a         = 0.37, delta_b           = -0.23, delta_c       = 0,
              lambda_LV_a = -0.11, lambda_LV_b = -0.1, lambda_LV_c = 0,
              # Alternative-specific variables
              b_MCH_a = 0.03, b_MCH_b = -0.5, b_MCH_c = 0.2,
              b_CAH_a = 0.3, b_CAH_b    = -1.6, b_CAH_c = 0.2,
              b_EM_a   = -0.6, b_EM_b     = -0.3, b_EM_c   = -0.2,
              b_PR_a   = -0.4, b_PR_b      = -0.3, b_PR_c   = -0.1,
              #Measurement equation parameters
              zeta_CD1 = 1, zeta_CD2 = 1,
              zeta_CD4 = 1, zeta_CD5 = 1,
              zeta_CD6 = 1, zeta_CD7 = 1,
              zeta_CD8 = 1, zeta_CD9 = 1,
              
              tau_CD1_1 = -3, tau_CD2_1 = -3,
              tau_CD1_2 = -2, tau_CD2_2 = -2,
              tau_CD1_3 = -1, tau_CD2_3 = -1,
              tau_CD1_4 =  1, tau_CD2_4 =  1,
              tau_CD1_5 =  2, tau_CD2_5 =  2,
              tau_CD1_6 =  3, tau_CD2_6 =  3,
              
              tau_CD4_1 = -3, tau_CD5_1 = -3,
              tau_CD4_2 = -2, tau_CD5_2 = -2,
              tau_CD4_3 = -1, tau_CD5_3 = -1,
              tau_CD4_4 =  1, tau_CD5_4 =  1,
              tau_CD4_5 =  2, tau_CD5_5 =  2,
              tau_CD4_6 =  3, tau_CD5_6 =  3,
              
              tau_CD6_1 = -3, tau_CD7_1 = -3,
              tau_CD6_2 = -2, tau_CD7_2 = -2,
              tau_CD6_3 = -1, tau_CD7_3 = -1,
              tau_CD6_4 =  1, tau_CD7_4 =  1,
              tau_CD6_5 =  2, tau_CD7_5 =  2,
              tau_CD6_6 =  3, tau_CD7_6 =  3,
              
              tau_CD8_1 = -3, tau_CD9_1 = -3,
              tau_CD8_2 = -2, tau_CD9_2 = -2,
              tau_CD8_3 = -1, tau_CD9_3 = -1,
              tau_CD8_4 =  1, tau_CD9_4 =  1,
              tau_CD8_5 =  2, tau_CD9_5 =  2,
              tau_CD8_6 =  3, tau_CD9_6 =  3
)

### 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("asc_PAYG", "delta_c", "lambda_LV_c")

# ################################################################# #
#### DEFINE RANDOM COMPONENTS                                    ####
# ################################################################# #

### Set parameters for generating draws
apollo_draws = list(
  interDrawsType = "MLHS", 
  interNDraws    = 500,
  interNormDraws = c("eta_LV")
)

### Create random parameters
apollo_randCoeff=function(apollo_beta, apollo_inputs){
  randcoeff = list()
  randcoeff[["LV"]] = eta_LV
  
  
  return(randcoeff)
}

### Read in starting values for at least some parameters from existing model output file
#apollo_beta = apollo_readBeta(apollo_beta, apollo_fixed, "MNL_SP", overwriteFixed=FALSE)

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

apollo_lcPars=function(apollo_beta, apollo_inputs){
  lcpars = list()
  
  lcpars[["asc_B1"]] = list(asc_B1_a, asc_B1_b, asc_B1_c)
  lcpars[["asc_B2"]] = list(asc_B2_a, asc_B2_b, asc_B2_c)
  lcpars[["asc_B3"]] = list(asc_B3_a, asc_B3_b, asc_B3_c)
  lcpars[["asc_B4"]] = list(asc_B4_a, asc_B4_b, asc_B4_c)
  
  lcpars[["b_MCH"]]  = list(b_MCH_a, b_MCH_b, b_MCH_c)
  lcpars[["b_CAH"]]  = list(b_CAH_a, b_CAH_b, b_CAH_c)
  lcpars[["b_EM"]]   = list(b_EM_a, b_EM_b, b_EM_c)
  lcpars[["b_PR"]]   = list(b_PR_a, b_PR_b, b_PR_c)
  
  ### Utilities of class allocation model
  V=list()
  V[["class_a"]]     = delta_a + lambda_LV_a*LV
  V[["class_b"]]     = delta_b + lambda_LV_b*LV
  V[["class_c"]]     = delta_c + lambda_LV_c*LV
  
  ### Settings for class allocation models
  classAlloc_settings = list(
    classes      = c(class_a=1, class_b=2, class_c=3),
    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()
  
  ### Likelihood of indicators
  ol_settings_CD1 = list(outcomeOrdered = CD1, #the column (variable) inside the database 
                         utility        = zeta_CD1*LV, #A numeric vector contains the explanatory var used in the OL
                         tau            = list(tau_CD1_1, tau_CD1_2, tau_CD1_3, tau_CD1_4, tau_CD1_5, tau_CD1_6), #A vector contains the threshold parameters (should have 1 element less than the Scale)
                         coding         = -3:3,
                         rows           = (choiceid==1),
                         componentName  = "indic_CD1")
  ol_settings_CD2 = list(outcomeOrdered = CD2, 
                         utility        = zeta_CD2*LV, 
                         tau            = list(tau_CD2_1, tau_CD2_2, tau_CD2_3, tau_CD2_4, tau_CD2_5, tau_CD2_6),
                         coding         = -3:3,
                         rows           = (choiceid==1),
                         componentName  = "indic_CD2")
  ol_settings_CD4 = list(outcomeOrdered = CD4, 
                         utility        = zeta_CD4*LV, 
                         tau            = list(tau_CD4_1, tau_CD4_2, tau_CD4_3, tau_CD4_4, tau_CD4_5, tau_CD4_6),
                         coding         = -3:3,
                         rows           = (choiceid==1),
                         componentName  = "indic_CD4")
  ol_settings_CD5 = list(outcomeOrdered = CD5, 
                         utility        = zeta_CD5*LV, 
                         tau            = list(tau_CD5_1, tau_CD5_2, tau_CD5_3, tau_CD5_4, tau_CD5_5, tau_CD5_6),
                         coding         = -3:3,
                         rows           = (choiceid==1),
                         componentName  = "indic_CD5")  
  ol_settings_CD6 = list(outcomeOrdered = CD6, 
                         utility        = zeta_CD6*LV, 
                         tau            = list(tau_CD6_1, tau_CD6_2, tau_CD6_3, tau_CD6_4, tau_CD6_5, tau_CD6_6),
                         coding         = -3:3,
                         rows           = (choiceid==1),
                         componentName  = "indic_CD6")
  ol_settings_CD7 = list(outcomeOrdered = CD7, 
                         utility        = zeta_CD7*LV, 
                         tau            = list(tau_CD7_1, tau_CD7_2, tau_CD7_3, tau_CD7_4, tau_CD7_5, tau_CD7_6),
                         coding         = -3:3,
                         rows           = (choiceid==1),
                         componentName  = "indic_CD7")  
  ol_settings_CD8 = list(outcomeOrdered = CD8, 
                         utility        = zeta_CD8*LV, 
                         tau            = list(tau_CD8_1, tau_CD8_2, tau_CD8_3, tau_CD8_4, tau_CD8_5, tau_CD8_6),
                         coding         = -3:3,
                         rows           = (choiceid==1),
                         componentName  = "indic_CD8")  
  ol_settings_CD9 = list(outcomeOrdered = CD9, 
                         utility        = zeta_CD9*LV, 
                         tau            = list(tau_CD9_1, tau_CD9_2, tau_CD9_3, tau_CD9_4, tau_CD9_5, tau_CD9_6),
                         coding         = -3:3,
                         rows           = (choiceid==1),
                         componentName  = "indic_CD9")  
  
  P[["indic_CD1"]] = apollo_ol(ol_settings_CD1, functionality)
  P[["indic_CD2"]] = apollo_ol(ol_settings_CD2, functionality)
  P[["indic_CD4"]] = apollo_ol(ol_settings_CD4, functionality)
  P[["indic_CD5"]] = apollo_ol(ol_settings_CD5, functionality)
  P[["indic_CD6"]] = apollo_ol(ol_settings_CD6, functionality)
  P[["indic_CD7"]] = apollo_ol(ol_settings_CD7, functionality)
  P[["indic_CD8"]] = apollo_ol(ol_settings_CD8, functionality)
  P[["indic_CD9"]] = apollo_ol(ol_settings_CD9, functionality)
  
  ### Combine Model
  P = apollo_combineModels(P, apollo_inputs, functionality)
  ### Take product across observation for same individual
  P = apollo_panelProd(P, apollo_inputs, functionality)
  ### Rename model
  names(P)[which(names(P)=="model")] <- "Measurement_model"
  
  ################################################################################################################################################
  ### Likelihood of choices inside each class
  P_within<-list()
  ### Loop over classes
  S = 3
  for(s in 1:S){
    
    ### Compute class-specific utilities
    ### Utilities for alternatives
    V = list()
    
    V[['PAYG']] = asc_PAYG
    V[['B1']] = asc_B1[[s]] + 
      b_MCH[[s]]*MCH_B1 +
      b_CAH[[s]]*CAH_B1 +
      b_EM[[s]]*EM_B1 +
      b_PR[[s]]*PR_B1
    V[['B2']] = asc_B2[[s]] +
      b_MCH[[s]]*MCH_B2 +
      b_CAH[[s]]*CAH_B2 +
      b_EM[[s]]*EM_B2 +
      b_PR[[s]]*PR_B2
    V[['B3']] = asc_B3[[s]] +
      b_MCH[[s]]*MCH_B3 +
      b_CAH[[s]]*CAH_B3 +
      b_EM[[s]]*EM_B3 +
      b_PR[[s]]*PR_B3
    V[['B4']] = asc_B4[[s]] + 
      b_MCH[[s]]*MCH_B4 +
      b_CAH[[s]]*CAH_B4 +
      b_EM[[s]]*EM_B4 +
      b_PR[[s]]*PR_B4
    
    ### Define settings for MNL model component
    mnl_settings = list(
      alternatives  = c(B1=1, B2=2, B3=3, B4=4, PAYG=5),
      avail         = list(B1=1, B2=1, B3=1, B4=1, PAYG=1), 
      choiceVar     = choice,
      utilities     = V
    )
    
    ### Compute within-class choice probabilities using MNL model
    P_within[[paste0("Class_",s)]] = apollo_mnl(mnl_settings, functionality)
    
    ### Take product across observation for same individual
    P_within[[paste0("Class_",s)]] = apollo_panelProd(P_within[[paste0("Class_",s)]], apollo_inputs ,functionality)
  }
  
  ### Compute latent class model probabilities
  lc_settings  = list(inClassProb = P_within, classProb=pi_values)
  
  P[["LC_model"]] = apollo_lc(lc_settings, apollo_inputs, functionality)
  
  ### Comment out as necessary
  P = apollo_combineModels(P, apollo_inputs, functionality)
  P = apollo_avgInterDraws(P, apollo_inputs, functionality)
  P = apollo_prepareProb(P, apollo_inputs, functionality)
  return(P)
}

# ################################################################# #
#### CALCULATE LL AT STARTING VALUES                             ####
# ################################################################# #

apollo_llCalc(apollo_beta, apollo_probabilities, apollo_inputs)

# ################################################################# #
#### MODEL ESTIMATION                                            ####
# ################################################################# #
### Optional starting values search
# apollo_beta=apollo_searchStart(apollo_beta, apollo_fixed,apollo_probabilities, apollo_inputs)
estimate_settings = list(estimationRoutine = "bfgs", maxIterations = 300)
model = apollo_estimate(apollo_beta, 
                        apollo_fixed, 
                        apollo_probabilities, 
                        apollo_inputs,
                        estimate_settings=estimate_settings)

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

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

apollo_modelOutput(model)
2. The output

Code: Select all

Preparing user-defined functions.

Testing likelihood function...
Apollo found a model component of type classAlloc without a componentName. The name was set to "classAlloc" by default.
INFORMATION: Setting "avail" is missing, so full availability is assumed. 

Overview of choices for OL model component indic_CD1:
                              1    2     3      4      5     6     7
Times chosen              25.00 4.00 31.00 134.00 129.00 77.00 88.00
Percentage chosen overall  5.12 0.82  6.35  27.46  26.43 15.78 18.03

Overview of choices for OL model component indic_CD2:
                              1     2     3      4     5     6     7
Times chosen              36.00 16.00 58.00 150.00 87.00 68.00 73.00
Percentage chosen overall  7.38  3.28 11.89  30.74 17.83 13.93 14.96

Overview of choices for OL model component indic_CD4:
                              1    2     3      4     5     6     7
Times chosen              57.00 21.0 91.00 168.00 55.00 42.00 54.00
Percentage chosen overall 11.68  4.3 18.65  34.43 11.27  8.61 11.07

Overview of choices for OL model component indic_CD5:
                              1     2     3      4     5     6    7
Times chosen              35.00 18.00 42.00 199.00 79.00 53.00 62.0
Percentage chosen overall  7.17  3.69  8.61  40.78 16.19 10.86 12.7

Overview of choices for OL model component indic_CD6:
                              1     2     3     4     5     6    7
Times chosen              46.00 18.00 48.00 182.0 78.00 54.00 62.0
Percentage chosen overall  9.43  3.69  9.84  37.3 15.98 11.07 12.7

Overview of choices for OL model component indic_CD7:
                              1     2     3      4   5     6     7
Times chosen              33.00 15.00 31.00 131.00 122 64.00 92.00
Percentage chosen overall  6.76  3.07  6.35  26.84  25 13.11 18.85

Overview of choices for OL model component indic_CD8:
                              1    2     3      4     5     6     7
Times chosen              64.00 20.0 75.00 169.00 59.00 44.00 57.00
Percentage chosen overall 13.11  4.1 15.37  34.63 12.09  9.02 11.68

Overview of choices for OL model component indic_CD9:
                              1     2     3      4     5     6     7
Times chosen              52.00 29.00 72.00 155.00 67.00 46.00 67.00
Percentage chosen overall 10.66  5.94 14.75  31.76 13.73  9.43 13.73
Apollo found a model component of type MNL without a componentName. The name was set to "Class_1" by default.

Overview of choices for MNL model component Class_1:
                                      B1      B2      B3      B4    PAYG
Times available                  2440.00 2440.00 2440.00 2440.00 2440.00
Times chosen                     1013.00  403.00  279.00  181.00  564.00
Percentage chosen overall          41.52   16.52   11.43    7.42   23.11
Percentage chosen when available   41.52   16.52   11.43    7.42   23.11

Apollo found a model component of type MNL without a componentName. The name was set to "Class_2" by default.

Overview of choices for MNL model component Class_2:
                                      B1      B2      B3      B4    PAYG
Times available                  2440.00 2440.00 2440.00 2440.00 2440.00
Times chosen                     1013.00  403.00  279.00  181.00  564.00
Percentage chosen overall          41.52   16.52   11.43    7.42   23.11
Percentage chosen when available   41.52   16.52   11.43    7.42   23.11

Apollo found a model component of type MNL without a componentName. The name was set to "Class_3" by default.

Overview of choices for MNL model component Class_3:
                                      B1      B2      B3      B4    PAYG
Times available                  2440.00 2440.00 2440.00 2440.00 2440.00
Times chosen                     1013.00  403.00  279.00  181.00  564.00
Percentage chosen overall          41.52   16.52   11.43    7.42   23.11
Percentage chosen when available   41.52   16.52   11.43    7.42   23.11

Apollo found a model component of type LC without a componentName. The name was set to "LC_model" by default.

Summary of class allocation for model component LC_model:
         Mean prob.
Class_1      0.4463
Class_2      0.2448
Class_3      0.3089
The class allocation probabilities for model component "LC_model" are calculated at the observation level in 'apollo_lcPars', but are used in 'apollo_probabilities' to multiply within class probabilities
  that are at the individual level. Apollo will average the class allocation probabilities across observations for the same individual level before using them to multiply the within-class probabilities.
  If your class allocation probabilities are constant across choice situations for the same individual, then this is of no concern. If your class allocation probabilities however vary across choice
  tasks, then you should change your model specification in 'apollo_probabilities' to only call 'apollo_panelProd' after calling 'apollo_lc'.

Pre-processing likelihood function...
Creating cluster...
Preparing workers for multithreading...

Testing influence of parameters....................................................................................
Starting main estimation
Initial function value: -9304.484 
Initial gradient value:
   asc_B1_a    asc_B1_b    asc_B1_c    asc_B2_a    asc_B2_b    asc_B2_c    asc_B3_a    asc_B3_b    asc_B3_c    asc_B4_a    asc_B4_b    asc_B4_c     delta_a     delta_b lambda_LV_a lambda_LV_b     b_MCH_a     b_MCH_b 
 619.253042   -8.012079  -13.117649 -186.906333  -10.777510  -58.457545 -225.532720  -14.284142   53.826554  -81.379951  -12.333778   25.793286   57.847063    3.834919  -44.698363   15.364139 -673.057417  -76.050374 
    b_MCH_c     b_CAH_a     b_CAH_b     b_CAH_c      b_EM_a      b_EM_b      b_EM_c      b_PR_a      b_PR_b      b_PR_c    zeta_CD1    zeta_CD2    zeta_CD4    zeta_CD5    zeta_CD6    zeta_CD7    zeta_CD8    zeta_CD9 
 141.089464 -471.063244  -43.142509  107.682175   93.581417  -23.833712   16.197255 -873.984638 -112.057753  239.968611   58.403859  106.289019  100.445859  110.788113   99.066008  110.699777   89.400626  131.526987 
  tau_CD1_1   tau_CD2_1   tau_CD1_2   tau_CD2_2   tau_CD1_3   tau_CD2_3   tau_CD1_4   tau_CD2_4   tau_CD1_5   tau_CD2_5   tau_CD1_6   tau_CD2_6   tau_CD4_1   tau_CD5_1   tau_CD4_2   tau_CD5_2   tau_CD4_3   tau_CD5_3 
  13.133116   14.061223  -20.717149  -25.483745  -30.109915   -2.838584  -97.673592  -35.333846    6.206948  -11.144011   -3.333438    6.444154   27.495607    9.823127  -41.116353  -11.570677   42.758658  -40.123730 
  tau_CD4_4   tau_CD5_4   tau_CD4_5   tau_CD5_5   tau_CD4_6   tau_CD5_6   tau_CD6_1   tau_CD7_1   tau_CD6_2   tau_CD7_2   tau_CD6_3   tau_CD7_3   tau_CD6_4   tau_CD7_4   tau_CD6_5   tau_CD7_5   tau_CD6_6   tau_CD7_6 
  22.330265   -1.070166    2.345656    2.784178    4.853262    6.731696   20.718624   10.447771  -14.987252   -7.385996  -25.984911  -25.017847   -8.243174  -90.087457    3.219968   16.834647    7.124427  -14.018746 
  tau_CD8_1   tau_CD9_1   tau_CD8_2   tau_CD9_2   tau_CD8_3   tau_CD9_3   tau_CD8_4   tau_CD9_4   tau_CD8_5   tau_CD9_5   tau_CD8_6   tau_CD9_6 
  35.486290   17.125129  -29.959341  -15.694015   19.605523   22.185395   13.596882   -1.686622    1.282722    4.178024    3.909245   -3.368930 
initial  value 9304.484190 
iter   2 value 8882.770889
iter   3 value 8256.357190
iter   4 value 8186.283191
iter   5 value 8016.243014
iter   6 value 7984.542712
iter   7 value 7913.618042
iter   8 value 7859.011997
iter   9 value 7812.047903
iter  10 value 7780.736819
iter  11 value 7736.691833
iter  12 value 7723.543087
iter  13 value 7712.871024
iter  14 value 7712.257706
iter  15 value 7699.591249
iter  16 value 7691.405840
iter  17 value 7680.866695
iter  18 value 7664.983329
iter  19 value 7651.417342
iter  20 value 7645.510078
iter  21 value 7636.885790
iter  22 value 7633.290422
iter  23 value 7629.007889
iter  24 value 7617.871190
iter  25 value 7611.425090
iter  26 value 7608.638876
iter  27 value 7603.326255
iter  28 value 7596.227317
iter  29 value 7586.068682
iter  30 value 7585.590213
iter  31 value 7584.352660
iter  32 value 7582.498257
iter  33 value 7579.471819
iter  34 value 7577.185032
iter  35 value 7574.739476
iter  36 value 7573.725708
iter  37 value 7572.470534
iter  38 value 7571.699454
iter  39 value 7570.854239
iter  40 value 7569.882826
iter  41 value 7569.516835
iter  42 value 7568.942007
iter  43 value 7568.463047
iter  44 value 7568.175006
iter  45 value 7567.767827
iter  46 value 7567.424527
iter  47 value 7567.223027
iter  48 value 7567.084437
iter  49 value 7566.882967
iter  50 value 7566.582124
iter  51 value 7566.518164
iter  52 value 7566.304149
iter  53 value 7566.071606
iter  54 value 7565.867347
iter  55 value 7565.611287
iter  56 value 7565.322904
iter  57 value 7565.156804
iter  58 value 7564.984536
iter  59 value 7564.778372
iter  60 value 7564.520533
iter  61 value 7564.185701
iter  62 value 7563.942224
iter  63 value 7563.802477
iter  64 value 7563.714958
iter  65 value 7563.640029
iter  66 value 7563.520927
iter  67 value 7563.377238
iter  68 value 7563.249434
iter  69 value 7563.163451
iter  70 value 7563.064432
iter  71 value 7562.968334
iter  72 value 7562.871095
iter  73 value 7562.820491
iter  74 value 7562.724303
iter  75 value 7562.635137
iter  76 value 7562.516082
iter  77 value 7562.431776
iter  78 value 7562.326540
iter  79 value 7562.236680
iter  80 value 7562.194514
iter  81 value 7562.113868
iter  82 value 7561.929829
iter  83 value 7561.792045
iter  84 value 7561.615635
iter  85 value 7561.568363
iter  86 value 7561.541420
iter  87 value 7561.493766
iter  88 value 7561.479203
iter  89 value 7561.444246
iter  90 value 7561.423128
iter  91 value 7561.416977
iter  92 value 7561.415191
iter  93 value 7561.412879
iter  94 value 7561.412720
iter  94 value 7561.412658
iter  94 value 7561.412656
final  value 7561.412656 
converged

Estimated parameters with approximate standard errors from BHHH matrix:
               Estimate     BHHH se BHH t-ratio (0)
asc_B1_a        5.60443     0.54299         10.3215
asc_B1_b       -0.91316     0.46736         -1.9539
asc_B1_c        0.35246     0.28270          1.2468
asc_B2_a        3.91355     0.56891          6.8791
asc_B2_b       -1.65458     0.81347         -2.0340
asc_B2_c        2.41741     0.20625         11.7205
asc_B3_a        2.02469     0.82184          2.4636
asc_B3_b        1.18095     1.79620          0.6575
asc_B3_c        2.18018     0.29319          7.4360
asc_B4_a        1.07005     1.24818          0.8573
asc_B4_b        2.60384     2.37347          1.0971
asc_B4_c        1.61577     0.36656          4.4079
asc_PAYG        0.00000          NA              NA
delta_a         0.42021     0.13676          3.0726
delta_b        -0.13113     0.14606         -0.8978
delta_c         0.00000          NA              NA
lambda_LV_a    -0.78729     0.13964         -5.6381
lambda_LV_b    -0.42306     0.17148         -2.4672
lambda_LV_c     0.00000          NA              NA
b_MCH_a         0.03296     0.24049          0.1371
b_MCH_b        -0.73826     0.61230         -1.2057
b_MCH_c         0.12785     0.09856          1.2972
b_CAH_a         0.24401     0.71344          0.3420
b_CAH_b        -2.53679     1.94914         -1.3015
b_CAH_c         0.14874     0.16491          0.9019
b_EM_a         -0.37514     0.12334         -3.0415
b_EM_b         -0.62732     0.32762         -1.9148
b_EM_c         -0.15226     0.10040         -1.5166
b_PR_a         -0.44137     0.36023         -1.2252
b_PR_b         -0.43282     1.00329         -0.4314
b_PR_c         -0.02598     0.09474         -0.2743
zeta_CD1        1.73521     0.16621         10.4398
zeta_CD2        2.71063     0.24262         11.1723
zeta_CD4        4.02563     0.29291         13.7434
zeta_CD5        4.45080     0.33454         13.3043
zeta_CD6        3.37225     0.26283         12.8305
zeta_CD7        2.28034     0.20470         11.1399
zeta_CD8        3.03796     0.23142         13.1276
zeta_CD9        3.73711     0.29985         12.4631
tau_CD1_1      -3.82554     0.45097         -8.4830
tau_CD2_1      -4.21309     0.45834         -9.1921
tau_CD1_2      -3.61678     0.39429         -9.1729
tau_CD2_2      -3.59625     0.35573        -10.1094
tau_CD1_3      -2.60596     0.23272        -11.1981
tau_CD2_3      -2.27057     0.25613         -8.8648
tau_CD1_4      -0.58073     0.16934         -3.4294
tau_CD2_4       0.03925     0.22257          0.1763
tau_CD1_5       0.97796     0.18127          5.3951
tau_CD2_5       1.57724     0.25253          6.2458
tau_CD1_6       2.25044     0.21779         10.3330
tau_CD2_6       3.36239     0.31461         10.6875
tau_CD4_1      -4.63144     0.40368        -11.4731
tau_CD5_1      -6.17810     0.55894        -11.0533
tau_CD4_2      -3.88816     0.37140        -10.4689
tau_CD5_2      -5.10061     0.51292         -9.9443
tau_CD4_3      -1.87276     0.29973         -6.2481
tau_CD5_3      -3.61908     0.38591         -9.3781
tau_CD4_4       1.42150     0.34622          4.1058
tau_CD5_4       0.51628     0.35316          1.4619
tau_CD4_5       3.31428     0.38480          8.6130
tau_CD5_5       2.78717     0.37108          7.5110
tau_CD4_6       5.60829     0.53533         10.4763
tau_CD5_6       5.51474     0.50693         10.8788
tau_CD6_1      -4.48758     0.38863        -11.5473
tau_CD7_1      -4.04256     0.36352        -11.1207
tau_CD6_2      -3.82043     0.35832        -10.6620
tau_CD7_2      -3.40600     0.32646        -10.4333
tau_CD6_3      -2.64323     0.28486         -9.2790
tau_CD7_3      -2.53423     0.25466         -9.9516
tau_CD6_4       0.46556     0.27178          1.7130
tau_CD7_4      -0.51553     0.20370         -2.5309
tau_CD6_5       2.37568     0.30524          7.7829
tau_CD7_5       1.20880     0.21168          5.7106
tau_CD6_6       4.45780     0.37423         11.9120
tau_CD7_6       2.47954     0.24694         10.0409
tau_CD8_1      -3.55840     0.35509        -10.0211
tau_CD9_1      -4.57605     0.44083        -10.3806
tau_CD8_2      -3.02587     0.32696         -9.2544
tau_CD9_2      -3.53322     0.36081         -9.7926
tau_CD8_3      -1.67057     0.26208         -6.3744
tau_CD9_3      -1.98351     0.29375         -6.7523
tau_CD8_4       1.00452     0.26754          3.7546
tau_CD9_4       0.82893     0.29193          2.8395
tau_CD8_5       2.52434     0.29321          8.6093
tau_CD9_5       2.65529     0.33193          7.9995
tau_CD8_6       4.34079     0.39952         10.8651
tau_CD9_6       4.49737     0.38483         11.6867

Final LL: -7561.4127


Summary of class allocation for model component LC_model:
         Mean prob.
Class_1      0.4474
Class_2      0.2454
Class_3      0.3073

Calculating log-likelihood at equal shares (LL(0)) for applicable models...
Calculating log-likelihood at observed shares from estimation data (LL(c)) for applicable models...
Calculating LL of each model component...
Calculating other model fit measures
INFORMATION: Your model took more than 10 minutes to estimate, so it was saved to file output/MNL_model.rds before calculating its covariance matrix. If calculation of the covariance matrix fails or is stopped before
  finishing, you can load the model up to this point using apollo_loadModel. You may also want to inspect the approximate BHHH standard errors shown above to determine whether you wish to continue this
  process. 
Computing covariance matrix using numerical methods (numDeriv).
 0%....25%....50%....75%....100% (332 NA values)
Computing covariance matrix using numerical methods (maxLik). This may take a while, no progress bar displayed.
WARNING: Some eigenvalues of the Hessian are positive, indicating convergence to a saddle point! 

Please acknowledge the use of Apollo by citing Hess & Palma (2019) - doi.org/10.1016/j.jocm.2019.100170
Warning message:
In sqrt(diag(varcov)) : NaNs produced
> 
> # model = apollo_estimate(apollo_beta, apollo_fixed, apollo_probabilities, apollo_inputs)
> 
> # ################################################################# #
> #### MODEL OUTPUTS                                               ####
> # ################################################################# #
> 
> # ----------------------------------------------------------------- #
> #---- FORMATTED OUTPUT (TO SCREEN)                               ----
> # ----------------------------------------------------------------- #
> 
> apollo_modelOutput(model)
Model run by Viet Hong Cung using Apollo 0.3.5 on R 4.3.2 for Windows.
Please acknowledge the use of Apollo by citing Hess & Palma (2019)
  DOI 10.1016/j.jocm.2019.100170
  www.ApolloChoiceModelling.com

Model name                                  : MNL
Model description                           : Simple MNL model MaaS bundle SP x LCCM 3C x LV
Model run at                                : 2025-11-04 17:07:10.861362
Estimation method                           : bfgs
Model diagnosis                             : successful convergence
Optimisation diagnosis                      : Saddle point found
     hessian properties                     : Some eigenvalues are positive and others negative
     maximum eigenvalue                     : 1393.792594
     reciprocal of condition number         : not calculated (Hessian is not negative definite)
Number of individuals                       : 488
Number of rows in database                  : 2440
Number of modelled outcomes                 : 6344
                       indic_CD1 : 488
                       indic_CD2 : 488
                       indic_CD4 : 488
                       indic_CD5 : 488
                       indic_CD6 : 488
                       indic_CD7 : 488
                       indic_CD8 : 488
                       indic_CD9 : 488
                        LC_model : 2440

Number of cores used                        :  16 
Number of inter-individual draws            : 500 (MLHS)

LL(start)                                   : -9304.48
LL (whole model) at equal shares, LL(0)     : -19120.69
LL (whole model) at observed shares, LL(C)  : -17175.4
LL(final, whole model)                      : -7561.41
Rho-squared vs equal shares                  :  0.6045 
Adj.Rho-squared vs equal shares              :  0.6001 
Rho-squared vs observed shares               :  0.5598 
Adj.Rho-squared vs observed shares           :  0.5577 
AIC                                         :  15290.83 
BIC                                         :  15642.81 

LL(0,indic_CD1)                  : -949.6
LL(final,indic_CD1)              : -817.96
LL(0,indic_CD2)                  : -949.6
LL(final,indic_CD2)              : -876.15
LL(0,indic_CD4)                  : -949.6
LL(final,indic_CD4)              : -874.19
LL(0,indic_CD5)                  : -949.6
LL(final,indic_CD5)              : -834.87
LL(0,indic_CD6)                  : -949.6
LL(final,indic_CD6)              : -857.13
LL(0,indic_CD7)                  : -949.6
LL(final,indic_CD7)              : -853.86
LL(0,indic_CD8)                  : -949.6
LL(final,indic_CD8)              : -876.01
LL(0,indic_CD9)                  : -949.6
LL(final,indic_CD9)              : -896.65
LL(0,Measurement_model)          : -7596.83
LL(final,Measurement_model)      : -5353.24
LL(0,LC_model)                   : -3927.03
LL(final,LC_model)               : -2229.41

Estimated parameters                        : 84
Time taken (hh:mm:ss)                       :  01:56:50.98 
     pre-estimation                         :  00:00:45.24 
     estimation                             :  00:30:47.21 
     post-estimation                        :  01:25:18.52 
Iterations                                  :  95  

Unconstrained optimisation.

Estimates:
               Estimate        s.e.   t.rat.(0)    Rob.s.e. Rob.t.rat.(0)
asc_B1_a        5.60443     0.96918      5.7826     2.71776        2.0622
asc_B1_b       -0.91316         NaN         NaN     0.60239       -1.5159
asc_B1_c        0.35246     0.39398      0.8946     0.96480        0.3653
asc_B2_a        3.91355     1.07472      3.6415     3.07004        1.2748
asc_B2_b       -1.65458     0.43014     -3.8466     1.63106       -1.0144
asc_B2_c        2.41741     0.29856      8.0969     0.65601        3.6850
asc_B3_a        2.02469     1.60529      1.2613     4.72546        0.4285
asc_B3_b        1.18095     0.43210      2.7330     2.52718        0.4673
asc_B3_c        2.18018     0.42529      5.1264     0.89089        2.4472
asc_B4_a        1.07005     2.08729      0.5126     6.03586        0.1773
asc_B4_b        2.60384     2.16542      1.2025     6.72301        0.3873
asc_B4_c        1.61577     0.51387      3.1443     1.03433        1.5621
asc_PAYG        0.00000          NA          NA          NA            NA
delta_a         0.42021     0.12735      3.2998     0.16817        2.4987
delta_b        -0.13113     0.12413     -1.0564     0.18012       -0.7280
delta_c         0.00000          NA          NA          NA            NA
lambda_LV_a    -0.78729     0.13120     -6.0006     0.14852       -5.3008
lambda_LV_b    -0.42306     0.13479     -3.1386     0.13103       -3.2288
lambda_LV_c     0.00000          NA          NA          NA            NA
b_MCH_a         0.03296         NaN         NaN     0.04434        0.7434
b_MCH_b        -0.73826     0.66394     -1.1119     1.84104       -0.4010
b_MCH_c         0.12785     0.05115      2.4993     0.09033        1.4154
b_CAH_a         0.24401     0.12246      1.9927     0.34840        0.7004
b_CAH_b        -2.53679     1.22832     -2.0652     2.66789       -0.9509
b_CAH_c         0.14874     0.10777      1.3802     0.16174        0.9196
b_EM_a         -0.37514     0.14418     -2.6018     0.27793       -1.3498
b_EM_b         -0.62732     0.30463     -2.0593     0.61422       -1.0213
b_EM_c         -0.15226     0.09103     -1.6727     0.11531       -1.3205
b_PR_a         -0.44137     0.22404     -1.9700     0.63499       -0.6951
b_PR_b         -0.43282     0.22798     -1.8985     1.35005       -0.3206
b_PR_c         -0.02598         NaN         NaN     0.10752       -0.2417
zeta_CD1        1.73521     0.14055     12.3456     0.16997       10.2088
zeta_CD2        2.71063     0.19939     13.5943     0.23784       11.3967
zeta_CD4        4.02563     0.30216     13.3230     0.41153        9.7820
zeta_CD5        4.45080     0.35187     12.6488     0.46206        9.6325
zeta_CD6        3.37225     0.24800     13.5980     0.32978       10.2258
zeta_CD7        2.28034     0.17485     13.0419     0.21536       10.5885
zeta_CD8        3.03796     0.22660     13.4067     0.30947        9.8167
zeta_CD9        3.73711     0.28196     13.2540     0.35507       10.5249
tau_CD1_1      -3.82554     0.24719    -15.4763     0.22220      -17.2170
tau_CD2_1      -4.21309     0.25660    -16.4187     0.24648      -17.0929
tau_CD1_2      -3.61678     0.23064    -15.6812     0.20662      -17.5044
tau_CD2_2      -3.59625     0.21876    -16.4393     0.22719      -15.8294
tau_CD1_3      -2.60596     0.17118    -15.2233     0.17931      -14.5335
tau_CD2_3      -2.27057     0.15938    -14.2467     0.17152      -13.2376
tau_CD1_4      -0.58073     0.11889     -4.8844     0.13636       -4.2587
tau_CD2_4       0.03925         NaN         NaN     0.08826        0.4447
tau_CD1_5       0.97796     0.13094      7.4688     0.18807        5.2001
tau_CD2_5       1.57724     0.14606     10.7986     0.17622        8.9506
tau_CD1_6       2.25044     0.17139     13.1303     0.24023        9.3680
tau_CD2_6       3.36239     0.24053     13.9793     0.30734       10.9404
tau_CD4_1      -4.63144     0.31897    -14.5198     0.41246      -11.2289
tau_CD5_1      -6.17810     0.44136    -13.9980     0.51492      -11.9981
tau_CD4_2      -3.88816     0.28183    -13.7960     0.37378      -10.4022
tau_CD5_2      -5.10061     0.36773    -13.8704     0.42101      -12.1152
tau_CD4_3      -1.87276     0.20831     -8.9905     0.29423       -6.3650
tau_CD5_3      -3.61908     0.29082    -12.4444     0.37192       -9.7309
tau_CD4_4       1.42150     0.21325      6.6659     0.30820        4.6122
tau_CD5_4       0.51628     0.18391      2.8072     0.27012        1.9113
tau_CD4_5       3.31428     0.28309     11.7073     0.39059        8.4853
tau_CD5_5       2.78717     0.27349     10.1910     0.40797        6.8319
tau_CD4_6       5.60829     0.42598     13.1655     0.58995        9.5064
tau_CD5_6       5.51474     0.44392     12.4229     0.60252        9.1528
tau_CD6_1      -4.48758     0.29648    -15.1363     0.39337      -11.4080
tau_CD7_1      -4.04256     0.27337    -14.7881     0.35949      -11.2452
tau_CD6_2      -3.82043     0.26146    -14.6118     0.34386      -11.1104
tau_CD7_2      -3.40600     0.23624    -14.4175     0.34417       -9.8962
tau_CD6_3      -2.64323     0.21349    -12.3813     0.29896       -8.8416
tau_CD7_3      -2.53423     0.19354    -13.0942     0.30428       -8.3288
tau_CD6_4       0.46556     0.16076      2.8960     0.21060        2.2106
tau_CD7_4      -0.51553     0.15331     -3.3626     0.32628       -1.5800
tau_CD6_5       2.37568     0.21581     11.0084     0.28491        8.3382
tau_CD7_5       1.20880     0.14786      8.1751     0.21905        5.5185
tau_CD6_6       4.45780     0.32374     13.7698     0.45579        9.7804
tau_CD7_6       2.47954     0.18675     13.2776     0.25318        9.7937
tau_CD8_1      -3.55840     0.24159    -14.7292     0.34242      -10.3919
tau_CD9_1      -4.57605     0.31373    -14.5860     0.36975      -12.3760
tau_CD8_2      -3.02587     0.21671    -13.9626     0.30473       -9.9298
tau_CD9_2      -3.53322     0.25893    -13.6457     0.31956      -11.0566
tau_CD8_3      -1.67057     0.17532     -9.5286     0.27001       -6.1870
tau_CD9_3      -1.98351     0.20436     -9.7058     0.28455       -6.9706
tau_CD8_4       1.00452     0.16976      5.9171     0.24176        4.1549
tau_CD9_4       0.82893     0.19804      4.1856     0.34023        2.4364
tau_CD8_5       2.52434     0.21465     11.7600     0.28289        8.9234
tau_CD9_5       2.65529     0.24170     10.9858     0.34687        7.6550
tau_CD8_6       4.34079     0.31664     13.7090     0.40637       10.6819
tau_CD9_6       4.49737     0.33850     13.2862     0.49667        9.0550


Summary of class allocation for model component LC_model:
         Mean prob.
Class_1      0.4474
Class_2      0.2454
Class_3      0.3073
I am also sharing the output of the simpler model without the Latent variable for your reference.

Code: Select all

Model run by Viet Hong Cung using Apollo 0.3.5 on R 4.3.2 for Windows.
Please acknowledge the use of Apollo by citing Hess & Palma (2019)
  DOI 10.1016/j.jocm.2019.100170
  www.ApolloChoiceModelling.com

Model name                                  : MNL
Model description                           : Simple MNL model MaaS bundle SP x LCCM 3C x LV
Model run at                                : 2025-11-04 17:07:10.861362
Estimation method                           : bfgs
Model diagnosis                             : successful convergence
Optimisation diagnosis                      : Saddle point found
     hessian properties                     : Some eigenvalues are positive and others negative
     maximum eigenvalue                     : 1393.792594
     reciprocal of condition number         : not calculated (Hessian is not negative definite)
Number of individuals                       : 488
Number of rows in database                  : 2440
Number of modelled outcomes                 : 6344
                       indic_CD1 : 488
                       indic_CD2 : 488
                       indic_CD4 : 488
                       indic_CD5 : 488
                       indic_CD6 : 488
                       indic_CD7 : 488
                       indic_CD8 : 488
                       indic_CD9 : 488
                        LC_model : 2440

Number of cores used                        :  16 
Number of inter-individual draws            : 500 (MLHS)

LL(start)                                   : -9304.48
LL (whole model) at equal shares, LL(0)     : -19120.69
LL (whole model) at observed shares, LL(C)  : -17175.4
LL(final, whole model)                      : -7561.41
Rho-squared vs equal shares                  :  0.6045 
Adj.Rho-squared vs equal shares              :  0.6001 
Rho-squared vs observed shares               :  0.5598 
Adj.Rho-squared vs observed shares           :  0.5577 
AIC                                         :  15290.83 
BIC                                         :  15642.81 

LL(0,indic_CD1)                  : -949.6
LL(final,indic_CD1)              : -817.96
LL(0,indic_CD2)                  : -949.6
LL(final,indic_CD2)              : -876.15
LL(0,indic_CD4)                  : -949.6
LL(final,indic_CD4)              : -874.19
LL(0,indic_CD5)                  : -949.6
LL(final,indic_CD5)              : -834.87
LL(0,indic_CD6)                  : -949.6
LL(final,indic_CD6)              : -857.13
LL(0,indic_CD7)                  : -949.6
LL(final,indic_CD7)              : -853.86
LL(0,indic_CD8)                  : -949.6
LL(final,indic_CD8)              : -876.01
LL(0,indic_CD9)                  : -949.6
LL(final,indic_CD9)              : -896.65
LL(0,Measurement_model)          : -7596.83
LL(final,Measurement_model)      : -5353.24
LL(0,LC_model)                   : -3927.03
LL(final,LC_model)               : -2229.41

Estimated parameters                        : 84
Time taken (hh:mm:ss)                       :  01:56:50.98 
     pre-estimation                         :  00:00:45.24 
     estimation                             :  00:30:47.21 
     post-estimation                        :  01:25:18.52 
Iterations                                  :  95  

Unconstrained optimisation.

Estimates:
               Estimate        s.e.   t.rat.(0)    Rob.s.e. Rob.t.rat.(0)
asc_B1_a        5.60443     0.96918      5.7826     2.71776        2.0622
asc_B1_b       -0.91316         NaN         NaN     0.60239       -1.5159
asc_B1_c        0.35246     0.39398      0.8946     0.96480        0.3653
asc_B2_a        3.91355     1.07472      3.6415     3.07004        1.2748
asc_B2_b       -1.65458     0.43014     -3.8466     1.63106       -1.0144
asc_B2_c        2.41741     0.29856      8.0969     0.65601        3.6850
asc_B3_a        2.02469     1.60529      1.2613     4.72546        0.4285
asc_B3_b        1.18095     0.43210      2.7330     2.52718        0.4673
asc_B3_c        2.18018     0.42529      5.1264     0.89089        2.4472
asc_B4_a        1.07005     2.08729      0.5126     6.03586        0.1773
asc_B4_b        2.60384     2.16542      1.2025     6.72301        0.3873
asc_B4_c        1.61577     0.51387      3.1443     1.03433        1.5621
asc_PAYG        0.00000          NA          NA          NA            NA
delta_a         0.42021     0.12735      3.2998     0.16817        2.4987
delta_b        -0.13113     0.12413     -1.0564     0.18012       -0.7280
delta_c         0.00000          NA          NA          NA            NA
lambda_LV_a    -0.78729     0.13120     -6.0006     0.14852       -5.3008
lambda_LV_b    -0.42306     0.13479     -3.1386     0.13103       -3.2288
lambda_LV_c     0.00000          NA          NA          NA            NA
b_MCH_a         0.03296         NaN         NaN     0.04434        0.7434
b_MCH_b        -0.73826     0.66394     -1.1119     1.84104       -0.4010
b_MCH_c         0.12785     0.05115      2.4993     0.09033        1.4154
b_CAH_a         0.24401     0.12246      1.9927     0.34840        0.7004
b_CAH_b        -2.53679     1.22832     -2.0652     2.66789       -0.9509
b_CAH_c         0.14874     0.10777      1.3802     0.16174        0.9196
b_EM_a         -0.37514     0.14418     -2.6018     0.27793       -1.3498
b_EM_b         -0.62732     0.30463     -2.0593     0.61422       -1.0213
b_EM_c         -0.15226     0.09103     -1.6727     0.11531       -1.3205
b_PR_a         -0.44137     0.22404     -1.9700     0.63499       -0.6951
b_PR_b         -0.43282     0.22798     -1.8985     1.35005       -0.3206
b_PR_c         -0.02598         NaN         NaN     0.10752       -0.2417
zeta_CD1        1.73521     0.14055     12.3456     0.16997       10.2088
zeta_CD2        2.71063     0.19939     13.5943     0.23784       11.3967
zeta_CD4        4.02563     0.30216     13.3230     0.41153        9.7820
zeta_CD5        4.45080     0.35187     12.6488     0.46206        9.6325
zeta_CD6        3.37225     0.24800     13.5980     0.32978       10.2258
zeta_CD7        2.28034     0.17485     13.0419     0.21536       10.5885
zeta_CD8        3.03796     0.22660     13.4067     0.30947        9.8167
zeta_CD9        3.73711     0.28196     13.2540     0.35507       10.5249
tau_CD1_1      -3.82554     0.24719    -15.4763     0.22220      -17.2170
tau_CD2_1      -4.21309     0.25660    -16.4187     0.24648      -17.0929
tau_CD1_2      -3.61678     0.23064    -15.6812     0.20662      -17.5044
tau_CD2_2      -3.59625     0.21876    -16.4393     0.22719      -15.8294
tau_CD1_3      -2.60596     0.17118    -15.2233     0.17931      -14.5335
tau_CD2_3      -2.27057     0.15938    -14.2467     0.17152      -13.2376
tau_CD1_4      -0.58073     0.11889     -4.8844     0.13636       -4.2587
tau_CD2_4       0.03925         NaN         NaN     0.08826        0.4447
tau_CD1_5       0.97796     0.13094      7.4688     0.18807        5.2001
tau_CD2_5       1.57724     0.14606     10.7986     0.17622        8.9506
tau_CD1_6       2.25044     0.17139     13.1303     0.24023        9.3680
tau_CD2_6       3.36239     0.24053     13.9793     0.30734       10.9404
tau_CD4_1      -4.63144     0.31897    -14.5198     0.41246      -11.2289
tau_CD5_1      -6.17810     0.44136    -13.9980     0.51492      -11.9981
tau_CD4_2      -3.88816     0.28183    -13.7960     0.37378      -10.4022
tau_CD5_2      -5.10061     0.36773    -13.8704     0.42101      -12.1152
tau_CD4_3      -1.87276     0.20831     -8.9905     0.29423       -6.3650
tau_CD5_3      -3.61908     0.29082    -12.4444     0.37192       -9.7309
tau_CD4_4       1.42150     0.21325      6.6659     0.30820        4.6122
tau_CD5_4       0.51628     0.18391      2.8072     0.27012        1.9113
tau_CD4_5       3.31428     0.28309     11.7073     0.39059        8.4853
tau_CD5_5       2.78717     0.27349     10.1910     0.40797        6.8319
tau_CD4_6       5.60829     0.42598     13.1655     0.58995        9.5064
tau_CD5_6       5.51474     0.44392     12.4229     0.60252        9.1528
tau_CD6_1      -4.48758     0.29648    -15.1363     0.39337      -11.4080
tau_CD7_1      -4.04256     0.27337    -14.7881     0.35949      -11.2452
tau_CD6_2      -3.82043     0.26146    -14.6118     0.34386      -11.1104
tau_CD7_2      -3.40600     0.23624    -14.4175     0.34417       -9.8962
tau_CD6_3      -2.64323     0.21349    -12.3813     0.29896       -8.8416
tau_CD7_3      -2.53423     0.19354    -13.0942     0.30428       -8.3288
tau_CD6_4       0.46556     0.16076      2.8960     0.21060        2.2106
tau_CD7_4      -0.51553     0.15331     -3.3626     0.32628       -1.5800
tau_CD6_5       2.37568     0.21581     11.0084     0.28491        8.3382
tau_CD7_5       1.20880     0.14786      8.1751     0.21905        5.5185
tau_CD6_6       4.45780     0.32374     13.7698     0.45579        9.7804
tau_CD7_6       2.47954     0.18675     13.2776     0.25318        9.7937
tau_CD8_1      -3.55840     0.24159    -14.7292     0.34242      -10.3919
tau_CD9_1      -4.57605     0.31373    -14.5860     0.36975      -12.3760
tau_CD8_2      -3.02587     0.21671    -13.9626     0.30473       -9.9298
tau_CD9_2      -3.53322     0.25893    -13.6457     0.31956      -11.0566
tau_CD8_3      -1.67057     0.17532     -9.5286     0.27001       -6.1870
tau_CD9_3      -1.98351     0.20436     -9.7058     0.28455       -6.9706
tau_CD8_4       1.00452     0.16976      5.9171     0.24176        4.1549
tau_CD9_4       0.82893     0.19804      4.1856     0.34023        2.4364
tau_CD8_5       2.52434     0.21465     11.7600     0.28289        8.9234
tau_CD9_5       2.65529     0.24170     10.9858     0.34687        7.6550
tau_CD8_6       4.34079     0.31664     13.7090     0.40637       10.6819
tau_CD9_6       4.49737     0.33850     13.2862     0.49667        9.0550


Summary of class allocation for model component LC_model:
         Mean prob.
Class_1      0.4474
Class_2      0.2454
Class_3      0.3073
I am looking forward to receiving your support.

Re: Latent Class Choice Model: NaNs produced

Posted: 29 Nov 2025, 13:36
by stephanehess
Hi

in general, using LVs in the class allocation model is empirically difficult and often leads to issues.

In your case, I would look at those parameters for which you do not obtain classical standard errors

Stephane