Rstan - error

[Chunlei Zhang]

Hi there,

I received an error after run a stan model. It would be great, if anyone could help me to figure out the reason. Thanks for your time and consideration.

The below is my OS and R version.

1. Windows 7 Enterprise, 64 bit Operating System;

2. R version: R version 3.3.3 (2017-03-06), 64 bit.

The error message is the below.

> fit <- stan(model_code = "stan_code", data = schools_dat, iter = 1000, chains = 4)

SYNTAX ERROR, MESSAGE(S) FROM PARSER:

ERROR at line 1

  1:    stan_code

       ^

PARSER EXPECTED: <model declaration (or perhaps an earlier block)>

Error in stanc(file = file, model_code = model_code, model_name = model_name,  : 

  failed to parse Stan model '9b0dc7c5007252e0174a4f88f6c3c089' due to the above error.

>

The R code is the below (this is an example from the web site, https://github.com/stan-dev/rstan/wiki/RStan-Getting-Started ).

library("rstan") 

rstan_options(auto_write = TRUE)

options(mc.cores = parallel::detectCores())

stan_code <- "

data {

  int<lower=0> J; // number of schools 

  real y[J]; // estimated treatment effects

  real<lower=0> sigma[J]; // s.e. of effect estimates 

}

parameters {

  real mu; 

  real<lower=0> tau;

  real eta[J];

}

transformed parameters {

  real theta[J];

  for (j in 1:J)

    theta[j] = mu + tau * eta[j];

}

model {

  target += normal_lpdf(eta | 0, 1);

  target += normal_lpdf(y | theta, sigma);

}

"

 

schools_dat <- list(J = 8, 

                    y = c(28,  8, -3,  7, -1,  1, 18, 12),

                    sigma = c(15, 10, 16, 11,  9, 11, 10, 18)

)

fit <- stan(model_code = "stan_code", data = schools_dat, iter = 1000, chains = 4)

Thanks again,

Peter

[Bob Carpenter]

"stan_code" is not a well-formed model. Did you meant
to say file="stan_code"? Usually we use the ".stan" suffix
on Stan files.

- Bob

> --
> You received this message because you are subscribed to the Google Groups "Stan users mailing list" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to stan-users+...@googlegroups.com.
> To post to this group, send email to stan-...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.

[Chunlei Zhang]

I changed the stan_code to scode, and use scode in the function like the below. Same errror message showed.

stan(model_code = "scode", data = schools_dat, iter = 1000, chains = 4) 


 Any new though would be appreciated. Should I install C++ compiler in order to execute Rstan function?

Thanks,

Chunlei

[Chunlei Zhang]

[Chunlei Zhang]

The code is an example from the web located at https://github.com/stan-dev/rstan/wiki/RStan-Getting-Started, I do not understand why it did not work on my computer.

[Bob Carpenter]

That is not an example from that page. There's no example
there using "model_code". The first example on that page
is this:

fit <- stan(file = '8schools.stan', data = schools_dat,

iter = 1000, chains = 4)

You'll see the argument is file and it's pointing to a file
containing a Stan program.

- Bob

[Jonah Gabry]

If you want to use the model_code argument to pass the Stan program as a string (instead of using a separate file) then you can't put the name of the Stan program string in quotes. So 

fit <- stan(model_code = "scode", data = schools_dat, iter = 1000, chains = 4) 

won't work. It needs to be 

fit <- stan(model_code = scode, data = schools_dat, iter = 1000, chains = 4) 

We definitely recommend using a file instead of a string, but that should at least work. 

Jonah

> To unsubscribe from this group and stop receiving emails from it, send an email to stan-users+unsubscribe@googlegroups.com.