Stan Errors

[Jo Lweny]

Hello Stan People,
Anyone kindly help out of this
I am trying to run a simple spatial model in rstan using the script attached.
Apparently I end up getting errors(which I don't understand) as shown below:





If so, you need to call increment_log_prob() with the log absolute determinant of the Jacobian of the transform.
Left-hand-side of sampling statement:
    rho_tr ~ gamma(...)

COMPILING THE C++ CODE FOR MODEL '2d2f9cabb7124e9b34547d4f6f8fe9fc' NOW.
cygwin warning:
  MS-DOS style path detected: C:/R/R-32~1.1/etc/x64/Makeconf
  Preferred POSIX equivalent is: /cygdrive/c/R/R-32~1.1/etc/x64/Makeconf
  CYGWIN environment variable option "nodosfilewarning" turns off this warning.
  Consult the user's guide for more details about POSIX paths:
    http://cygwin.com/cygwin-ug-net/using.html#using-pathnames
[20,1,4]
[10,1,4]
Error in eval(expr, envir, enclos) : Exception thrown at line 21:
[]: accessing element out of range. index 1 out of range; expecting index to be between 1 and 0; index position = 1x
failed to create the sampler; sampling not done



Session Info(IF NECESSARY)

> sessionInfo() R version 3.2.1 (2015-06-18) Platform: x86_64-redhat-linux-gnu (64-bit) Running under: CentOS release 6.7 (Final) locale: [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C LC_TIME=C LC_COLLATE=C LC_MONETARY=C LC_MESSAGES=C LC_PAPER=C LC_NAME=C LC_ADDRESS=C LC_TELEPHONE=C LC_MEASUREMENT=C [12] LC_IDENTIFICATION=C attached base packages: [1] parallel tcltk grid splines stats graphics grDevices utils datasets methods base other attached packages: [1] stringr_1.0.0 geostatsp_1.3.5 SpatialExtremes_2.0-2 geoR_1.7-5.1 rstan_2.7.0-1 inline_0.3.14 Rcpp_0.11.1 RColorBrewer_1.1-2 foreign_0.8-63 raster_2.4-18 [11] R.basic_0.53.0 abind_1.4-3 fields_8.2-1 maps_2.3-9 spam_1.0-1 epiR_0.9-62 survival_2.38-1 maptools_0.8-36 lattice_0.20-31 INLA_0.0-1392038736 [21] Matrix_1.2-1 sp_1.0-14 R.utils_1.29.8 R.oo_1.19.0 R.methodsS3_1.7.0 loaded via a namespace (and not attached): [1] magrittr_1.5 MASS_7.3-40 RandomFields_3.1.1 tools_3.2.1 RandomFieldsUtils_0.0.10 numDeriv_2014.2-1 codetools_0.2-11 splancs_2.01-37 stringi_0.5-5 [10] stats4_3.2.1

>

[Daniel Lee]

Hi,

On Tue, Sep 8, 2015 at 9:22 AM, Jo Lweny <jolwe...@gmail.com> wrote:

Apparently I end up getting errors(which I don't understand) as shown below:

[20,1,4]
[10,1,4]
Error in eval(expr, envir, enclos) : Exception thrown at line 21:
[]: accessing element out of range. index 1 out of range; expecting index to be between 1 and 0; index position = 1x
failed to create the sampler; sampling not done

This is telling you that you have an error at line 21. It should be this line:
  for(i in 1:N_obs) x[i]<- x_obs[i];

It helps to put your Stan program into a separate file so you can easily track line numbers.

I'm guessing that x[i] and x_obs[i] are different sizes, but I don't know why it'd give you that message. I'll try to hunt it down what that message means.

[Jan]

row_vector [D] x_obs[N_obs];
row_vector[D] x[N];

Assigning one to the other would only work if their dimensions are the same, i.e. N==N_obs.

[Daniel Lee]

Ah. It's making sense now. I don't know if there's a better way to form that error message.

Stan is an imperative language (it executes things in the order you write it).

If you look at the transformed data block:
transformed data{
  int<lower=1> N;
  row_vector[D] x[N];
  matrix [N,D] x_mat;
  vector[N] mu;
  N<- N_obs + N_pred;
  ...

you are trying to declare x as: row_vector[D] x[N]. Here, N is still 0. It only gets set to N_obs + N_pred after x is declared.

One way to fix it is to declare x as:
  row_vector[D] x[N_obs + N_pred];

With that fix, you'll run into problems for x_mat and mu, but I'll let you take care of that.

Daniel

[Bob Carpenter]

Do we have a simple example to recreate this message? If so,
can someone create an issue to create a better formulated error
message?

I can't make heads or tails of the text of that message.

I don't like that "expecting index to be between 1 and 0" --- I
have no idea what it means and I may have written the error message.
And why is the position "1x" rather than "1"?

And why does it say "eval(expr, envir, enclos)" --- is that some
kind of carryover from R or something else?

- 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.
> For more options, visit https://groups.google.com/d/optout.

[Ben Goodrich]

On Tuesday, September 8, 2015 at 3:14:06 PM UTC-4, Bob Carpenter wrote:

And why does it say "eval(expr, envir, enclos)"  --- is that some
kind of carryover from R or something else?

R

Ben