I have a model which requires input data row_vector[K] x[T]. There is one such row_vector[,] per data sample, and there are N data samples. So, basically I have an array[N} of array[T] of array[K]. How do I feed this using a data.R file? is it

list(list(list( , , , ,))) or

c(c(c( , , , ,)

or something else?

`array(, dims = c(N, T, K))`

I am not asking about declaration, but how to feed the data in a file like model.data.R. With lines like

x <- …

how do I fill in the right side, replacing the …

By naming such an array in a call to `rstan::stan_rdump`

, which ultimately calls

```
if (is.matrix(vv) || is.array(vv)) {
l2 <- c(l2, v)
vvdim <- dim(vv)
cat(v, " <- \\n", file = file, sep = "")
if (length(vv) == 0) {
str <- paste0("structure(integer(0), ")
}
else {
str <- paste0("structure(c(", paste(as.vector(vv),
collapse = ", "), "),")
}
str <- gsub(addnlpat, "\\\\1\\n", str)
cat(str, ".Dim = c(", paste(vvdim, collapse = ", "),
"))\\n", file = file, sep = "")
next
}
```

I think I was being too brief. I am following the method shown in Bernoulli example, where you enter the command

examples/bernoulli/bernoulli sample data file=examples/bernoulli/bernoulli.data.R

and the bernoulli.data.R content is

N <- 10

y <- c(0,1,0,0,0,0,0,0,0,1)

Can I use the same approach by defining x <- in an data.R file?

You could, but it is easier to use `rstan::stan_rdump`

.

I read the stan_rdump manual. It is expecting an R object as input and will dump out the content to a file. I don’t have an R object to start with, so I am jumping thru more hoops.

Maybe if you tell me this. What is an R object that is equivalent to row_vector[K] x[T], I will create such R object, do a dump, and see what I get.

I think its dim

T <- 60

K <- 5

b <- array(1:300, dim = c(T, K))

it works up to here. then

stan_rdump(b, file="") gets

Error in FUN(X[[i]], …) : invalid first argument

`stan_rdump("b", file = "")`

yeah, I just got that too. Thanks.