SampleNMM: Normal-Uniform mixture model fitting

In CHAT: Clonal Heterogeneity Analysis Tool

Description

This function fits the distribution of a input vector with mixture of Normal and Uniform distributions and returns the BIC score for the model.

Usage

SampleNMM(Y)

Arguments

Y	input vector. In the context of subclonality inference, Y can either be sAGP or CCF values from one sample.

Details

The model is: Y~ A*U(0,1)+(1-A)*Normal(mu,sigma), where A, mu and sigma are parameters to be fitted.

Value

a list containing the following elements:

BIC	BIC score from the model.
fit	a vector containing three numbers, in the order of fitted A, mu and sigma.

Author(s)

Bo Li

Examples

Y=c(rnorm(100,0.5,0.3),runif(100,0,1))
SampleNMM(Y)

Example output

Loading required package: parallel
Loading required package: DNAcopy
Loading required package: DPpackage
 
DPpackage 1.1-7.4
 
Copyright (C) 2006 - 2012, Alejandro Jara
Department of Statistics
P.U. Catolica de Chile
 
Support provided by Fondecyt
11100144 grant.
 
$BIC
[1] 114.4992

$fit
[1] 0.03322251 0.52850711 0.30161807

Warning message:
In log(A/(max(y, na.rm = TRUE) - min(y, na.rm = TRUE)) + (1 - A) *  :
  NaNs produced

CHAT documentation built on May 29, 2017, 10:32 p.m.