nimbleMacros (version 0.1.1)

LM: Macro for fitting linear models, GLMs, and GLMMs

Description

This macro generates code for LMs, GLMs, and GLMMs using formula notation and arguments similar to R functions such as lm(), glm(), and lmer()/glmer(). Currently only normal, Poisson, and binomial models are supported.

Value

NIMBLE code for the linear model, GLM, or GLMM specified by the formula, including priors.

Arguments

formula

An R formula, possibly with the parameters followed by brackets containing indices. If there are no indices, the macro attempts to guess the correct indices from the context. Formulas can include random effects via lme4-style notation (e.g., ~ x + (1|group))

family

A description of the error distribution and link function to be used in the model. This can be a character string naming a family function, a family function or the result of a call to a family function. See ?family. Supported families are gaussian (default), binomial, and poisson.

coefPrefix

Character. All model coefficient names will begin with this prefix. default is "beta_" (so 'x' becomes 'beta_x', etc.)

sdPrefix

Character. All dispersion parameters will begin with this prefix. default is no prefix.

priors

List of prior specifications, generated using setPriors().

modelMatrixNames

Logical indicating if parameters be named so they match the names one would get from R's model.matrix.

modelInfo

Used internally by nimbleMacros; a list of model information such as constants and dimensions

.env

Used internally by nimbleMacros; the environment where the model was created

Author

Ken Kellner

Examples

constants <- list(y = rnorm(10),
                  x = rnorm(10), 
                  x2 = factor(sample(letters[1:3], 10, replace=TRUE)))
 
code <- nimbleCode({
   LM(y ~ x + x2)
})

mod <- nimbleModel(code, constants = constants)
mod$getCode()