EstimateTruncatedDictionary: Estimate Dictionary of Truncated Kernels

Description Usage Arguments Value Author(s) References

View source: R/EstimateTruncatedDictionary.R

Description

Given a matrix with methylation sites as rows and samples as columns, returns a collection of Gaussian kernels truncated between 0 and 1. These kernels are used as a dictionary, where the distribution of values at each site is represented a weighted combination of the kernels.

Usage

1
EstimateTruncatedDictionary(X, K=2, a0 = 0.5,b0 = 0.5,mu0 = 0.5,Concentration = 0.5, NumDraws = 1000)

Arguments

X

A matrix in which rows represent variables (e.g., methylation sites) and columns represent samples. The entries of the matrix must be continuous between 0 and 1.

K

The number of kernels

a0,b0

Gamma hyperperameters for the precision (inverse of the variance) of the kernels. Defaults are a0=0.5, b0=0.5 (we recommend using these defaults if unsure).

mu0

Normal mean hyperparameter for the kernel means. Default is mu0=0.5.

Concentration

Initial value of the Dirichlet concentration parameter. Default is 0.5. Can be a vector of length K.

NumDraws

Number of MCMC draws for posterior inference.

Value

Returns an object with the following values, averaged over the MCMC iterations:

Mu

Vector of length K giving the mean of each kernel

Sigma

Vector of length K giving the standard devitation of each kernel

Concentration

Vector of length K giving the concentration hyperparameter

Author(s)

Eric F. Lock

References

Lock, E. F. & Dunson, D. B. (2015). Shared kernel Bayesian screening. Biometrika, 102 (4): 829-842.


lockEF/BayesianScreening documentation built on May 24, 2020, 11:50 p.m.