Allow dispersion parameter to be observation-specific
Open
Nobody has claimed this yet.
enhancement
- Dominant language
- R
- Stars
- 114
- Forks
- 31
- PR merge metrics
- No merged PRs in 30d
Description
This is a feature request to allow the dispersion parameter to be observation-specific, as, for example, in models like the following:
library(brms)
set.seed(4734)
data_het <- data.frame(
y = c(rnorm(50), rnorm(50, 1, 2)),
x = factor(rep(c("a", "b"), each = 50)),
se_y = rgamma(100, shape = 1)
)
fit6 <- brm(y | se(se_y, sigma = TRUE) ~ x, data = data_het)
stancode(fit6)
str(standata(fit6))
This came up in https://discourse.mc-stan.org/t/accounting-for-measurement-error-during-variable-selection-with-projpred-possibly-with-rstanarm-or-brms/11789.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
The issue names no project files or tests; begin by tracing how brms models using se(se_y, sigma = TRUE) are handled. Use the supplied data_het example with stancode() and standata() to understand the current behavior. Done means observation-specific dispersion is supported for this model pattern and the generated Stan code and data reflect it.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- r
- Domain
- data
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100