Dynamic Logistic Model for Network Formation [Part1. Problems with data types]

Dear All, I need help implementing a logistic model for dynamic network formation. This is how I am passing the data from R.

#### DATA
dat.stan <-list(
  ## Indixes
  N=m, # Index of individuals.
  G=G, # Index of networks.
  P=P, # Index of time periods.
  W=W, # Array of binary elements (adjacency matrices NxN) for each network (G) and time period (P) W[N,N,G,P]
  C=C, # Array of binary elements (adjacency matrices NxN) for each network (G) and time period (P) that describe simmilarity between two individuals If c(i=j)=1, or else c(i diff j)=0 C[N,N,G,P]
  CC=CC # Array of characteristics of each individual in each network in each time period CC[N,G,P]

#### STAN
rstan_options(auto_write = T)
options(mc.cores = 3L)
# model options
file <- "network_formation.stan"
inter <- 100L
chains <- 3L
model_name <- "network_formation.stan"
init <-  0

fit <-
    file = file,
    data = dat.stan,
    iter = inter,
    chains = chains,
    model_name = model_name,
    cores = 3L,
    init = init

data {
  //// Indixes ////
  int<lower=0> N; // Index of units.
  int<lower=0> G; // Index of networks.
  int<lower=0> P; // Index of time periods.
  //// Data ////
  int<lower=0> W[N, N, G, P]; // array of G adjacency matrices dimension NxN for each P period
  real<lower=1> C=[N,N,G,P]; // space for dyad specific variable in network formation
  real CC=[N,G,P]; // space for individual specific variable in network formation

parameters {
  real gamma1;
  real gamma2;
  real gamma3;
  real gamma4;
model {
  for(t in 1:P){
    for(g in 1:G){
      for(i in 1:N){
        for(j in 1:N){
        W[i, j, g, t] ~ bernoulli_logit(gamma1 + gamma2 * C[i, j, g, t] + gamma[3] * CC[i, g, t] + gamma[4] * CC[j, g, t]);


I am getting the following error

Variable definition not possible in this block.
Variable definition base type mismatch, variable declared as base type real variable definition has base type row_vectorVariable definition dimensions mismatch, definition specifies 0, declaration specifies 1 error in ‘model4188be481_network_formation_stan’ at line 11, column 27

1 Like

I guess the extra = after C is probably the culprit.

Best of luck with the model.


Oh, I feel so dummy jejejej. TX a lot.

No worries :-) Glad you are able to move forward.