kmm - 1 year ago 42

R Question

, which has a functions and to carry out Bayesian estimation (the latter as an interface to ). and take an of formulas to define the likelihood and priors, such as:

which prints as:

I'm trying to explore the effects of priors, so I want to be able to pass different sets of means and standard deviations. I can get the priors into a semicolon separated string like:

from which I think I should be able to split on and assign to an . But I can't figure out this last step.

I have tried variations of:

and

The former fails with and the latter because doesn't have an "alist" mode.

Maybe the more general question is: how can I get a string into an? I have read the answer in this question, but I can't figure out how to adapt it for this particular problem.

`map()`

`map2stan()`

`rstan`

`map()`

`map2stan()`

`alist()`

`alist(`

height ~ dnorm(mu, sigma),

mu ~ dnorm(178, 20),

sigma ~ dunif(0, 50)

)

which prints as:

`[[1]]`

height ~ dnorm(mu, sigma)

[[2]]

mu ~ dnorm(178, 20)

[[3]]

sigma ~ dunif(0, 50)

I'm trying to explore the effects of priors, so I want to be able to pass different sets of means and standard deviations. I can get the priors into a semicolon separated string like:

`s <- "height ~ dnorm(mu, sigma); mu ~ dnorm(178, 20); sigma ~ dunif(0, 50)"`

from which I think I should be able to split on

`;`

`alist()`

I have tried variations of:

`do.call("alist", unlist(strsplit(s,";")))`

and

`as.vector(unlist(strsplit(s,";")), mode = "alist")`

The former fails with

`second argument must be a list`

`as.vector()`

Maybe the more general question is: how can I get a string into an

`alist()`

Answer Source

This will do the job:

```
lapply(strsplit(s, ";")[[1]], function(x) parse(text = x)[[1]])
```