miniSim function

Mini Simulator

Function to do a miniature simulation testing the various grouped stagewise techniques.

miniSim(sampleSize = 50, clusterSize = 4, responseCor = 0.4,
  groupSparsity = "No Sparsity", xVariance = 1, covariateCor = 0.4,
  intercept = 1, SNR = 10, techniques = c("SEE", "BiSEE", "GSEE",
  "HiSEE"), stepSize = 0.125, maxNumSteps = 400, reps = 200)

Arguments

• sampleSize: Number of clusters to be used in each replicate.
• clusterSize: Size of each cluster.
• responseCor: Correlation parameter for response.
• groupSparsity: String specifying within group sparsity level; currently takes values 'No Sparsity', 'Mod Sparsity', or 'High Sparsity'.
• xVariance: Scalar value indicating marginal variance of the covariates.
• covariateCor: Correlation parameter used for within group correlation.
• intercept: Scalar value indicating the true intercept value.
• SNR: Optional scalar value that allows fixing the signal to noise ratio as defined as the ratio of the variance in the linear predictor to the variance of the response conditioned on the covariates.
• techniques: Vector of all techniques to be used for comparison; default value of c("SEE", "BiSEE", "GSEE", "HiSEE") currently contains all possible values.
• stepSize: Stepsize to be used in stagewise techniques.
• maxNumSteps: Maximum number of steps to be taken by stagewise techniques.
• reps: Number of replicates to be used for simulation.

Returns

Simulation results.

Details

Function is currently under development and is thus not exported. Function is intended to allow for simple simulation implementation to explore the provided stagewise regression functions.

Note

Currently still under development, so not exported.

Author(s)

Gregory Vaughan