bvs_task: Variational Bayes Variable Selection Problem

In bhtang127/AccelBenchmark: Benchmark Various Accelerating Methods for Fixed Point Iterations

Variational Bayes Variable Selection Problem

Description

A function that will generate the task list for variational Bayes variable selection problem. Once ran, it will generate a simulated data that logit(p_i) = -1 - Z_{i1} - Z_{i2} + X_i^T β, where Z_i1, Z_i2 are independent standard normal variable. β are i.i.d follow Normal(0, 0.25) but some of them are set to be zero X where drawn independently from Binomial(2, p) where p is uniform over (0.05, 0.5)

Usage

bvs_task(n = 400, p = 1000, sd = 0.1, rate = 0.1)

Arguments

n	Number of rows in simulated data
p	Number of predictors in simulated data
sd	Standard deviation used in random initialize function
rate	Only floor(rate * p) components in β will be non-zero

Value

A list containing all components needed for benchmarking the problem

initfn	Parameter random initializing function
fixptfn	Updating function for the fixed point iteration problem
objfn	Function calculating the objective value for current parameter
...	Other arguments required in functions above

References

Carbonetto P, Stephens M, et al. (2012). Scalable variational inference for Bayesian variable selection in regression, and its accuracy in genetic association studies. Bayesian Analysis, 7(1): 73–108.

Examples

## Not run: 
set.seed(54321)
problem = bvs_task(n=200, p=2000, rate=0.1)
benchmark(
  problem,
  algorithm=c("raw", "squarem", "daarem", "pem", "qn", "nes"),
  ntimes = 200
)

## End(Not run)

bhtang127/AccelBenchmark documentation built on May 30, 2022, 2:21 a.m.