confint.em: Confint method of S3-class 'em'.

In qmiao19/CytoComp: de novo CyTOF compensation

Description

This method of the S3 em class calculates the confidence intervals of the parameters of the fitted finite mixture model.

Usage

## S3 method for class 'em'
confint(object, t = 1e-64, nb = 10, level = 0.95)

Arguments

t	A numerical scalar indicating the value below which the E-M algorithm should stop.
nb	Number of Monte Carlo simulations.
level	The confidence level required.

Details

Confidence intervals of the parameters of probability distributions are calculated by the confint method of the S4 confint class of the bbmle package with default values. See the help of this method for technical details. The confidence interval of the lambda parameter is calculated thanks to the information-based method of Oakes (1999). The possible non-independance between lambda and the parameters of the probability distributions is accounted for by Monte Carlo simulations where each iteration consists in (i) sampling values of theses parameters in a multinormal distribution, and (ii) applying the method of Oakes (1999). Samplings in the multinormal distribution is performed by the rmultinormal function of the mc2d package.

Value

A dataframe containing point estimates (first column) and confidence intervals (second and third columns) of each of the five parameters of the finite mixture model (five row, one per parameter).

References

David Oakes (1999) Direct calculation of the information matrix via the EM algorithm. J R Statist Soc B, 61: 479-482.

qmiao19/CytoComp documentation built on Nov. 5, 2019, 1:57 a.m.