Single scenario simulation

[zywhy9]

`library(simanalyse)
library(simcases)
library(coda)

set.seed(123)

lik <- "
Ew ~ dbin(p,N) # Weir data
p ~ dbeta(a,b) # GSI proportion
"

sd <- runif(1,0,0.1) ## standard deviation of p given by GSI
mu <- runif(1) ## mean of p given by GSI
rho <- mu*(1-mu)/sd^2-1
a <- rho*mu
b <- rho-a

const = list(N=500)
sigma = list(a=a,b=b)
all = "."

models_sims = "code constants parameters monitor
lik const sigma all"
smc_simulate(models = models_sims,
nsims = 1,
exists = NA,
ask = FALSE)`

Error in names(list_model) <- model_header :
'names' attribute [4] must be the same length as the vector [1]

[joethorley]

What is the traceback?

[audrey-b]

There seems to be a problem when models_sims contains only one model, but adding a second model fixes it. I'll have to look into it. zywhy9 is my PhD student so you might see him post issues here and there.

[zywhy9]

Sorry, I'm not sure what the traceback means. The output is

> library(simanalyse)
> library(simcases)
> library(coda)
> set.seed(123)
> 
> lik <- "
+   Ew ~ dbin(p,N)           # Weir data
+   p ~ dbeta(a,b)           # GSI proportion
+ "
> 
> ## Simulated Data
> sd <- runif(1,0,0.1) ## standard deviation of p given by GSI
> mu <- runif(1) ## mean of p given by GSI
> rho <- mu*(1-mu)/sd^2-1
> a <- rho*mu 
> b <- rho-a
> 
> N = nlist(N=500)
> const = nlist(a=a,b=b)
> all = "."
> models_sims = "code constants parameters monitor
+                lik  const     N          all"
> smc_simulate(models = models_sims,
+              nsims = 1,
+              exists = NA,
+              ask = FALSE)
Error in names(list_model) <- model_header : 
  'names' attribute [4] must be the same length as the vector [1]

[joethorley]

immediately after you receive the error type

traceback()

By default traceback() prints the call stack of the last uncaught error, i.e., the sequence of calls that lead to the error. This is useful when an error occurs with an unidentifiable error message.

[zywhy9]

it gives:

2: args_to_list(models = models, environment = environment, fun = fun)
1: smc_simulate(models = models_sims, nsims = 1, exists = NA, ask = FALSE)

[joethorley]

OK - I was just worried it was a problem with sims.