map_hyperprior_elicitation: Evaluate the approximation in the hyperprior elicitation for...

In hruffieux/atlasqtl: Flexible sparse regression for hierarchically-related responses using annealed variational inference

Evaluate the approximation in the hyperprior elicitation for the number of predictors associated with each response

Description

This function is used to guide hyperprior elicitation before running atlasqtl. The errors arising from the approximation in the hyper prior elicitation are estimated by Monte-Carlo simulation.

Usage

map_hyperprior_elicitation(p0, p, q, n_draws = 1e+05, n_cpus = 1)

Arguments

p0	Vector of size 2 whose entries are the prior expectation and variance of the number of predictors associated with each response.
p	Number of candidate predictors.
q	Number of responses.
n_draws	Number of draws used for Monte-Carlo simulation (default 1e5).
n_cpus	Number of cores used (default serial).

Value

A list of errors for the prior mean and prior standard deviation of the number of predictors associated with each response.

See Also

atlasqtl

Examples

seed <- 123; set.seed(seed)
n <- 200; p <- 100; q <- 20000
p0 <- c(1, 10)
n_draws <- 1e5 # must be large for accurate estimation

map_hyperprior_elicitation(p0, p, q, n_draws, n_cpus = 1)

hruffieux/atlasqtl documentation built on April 12, 2025, 12:54 p.m.