FGH: The SCIs Based on FGH Methods

In DataXujing/LN0SCIs: Simultaneous CIs for Ratios of Means of Log-Normal Populations with Zeros

Description

A method based on generalized pivotal quantity with order statistics(also see FGW) to construct the simultaneous confidence intervals for Ratios of Means of Log-normal Populations with excess Zeros.

Usage

FGH(n,p,mu,sigma,N,C2=rbind(c(-1,1,0),c(-1,0,1),c(0,-1,1)),alpha=0.05)

Arguments

p	The zero probability of the mixture distribution,it has the same length to the n params.
N	The number of independent generated data sets.
n	The sample size of the mixture distributions,must be an integer vector.
mu	The mean of the non-zero samples,which after log-transformation.
sigma	The standard deviation of the non-zero samples,which after log-transformation.
C2	Matrix C,You can refer to the paper of Xu et al. for specific forms.
alpha	The confidence level,it always set alpha=0.5

Details

More information about FGH, you can read the paper: Simultaneous Confidence Intervals for Ratios of Means of Log-normal Populations with Zeros.

Value

The method will return the Simultaneous Confidence Intervals(SCIs) and the time consuming

Author(s)

Jing Xu, Xinmin Li, Hua Liang

Examples

alpha <- 0.05

p <- c(0.1,0.15,0.1)
n <- c(30,15,50)
mu <- c(0,0,0)
sigma <- c(1,1,1)
N <- 500
FGH(n,p,mu,sigma,N)

## Not run: 
p <- c(0.1,0.15,0.1,0.6)
n <- c(30,15,10,50)
mu <- c(0,0,0,0)
sigma <- c(1,1,1,1)
C2 <- rbind(c(-1,1,0,0),c(-1,0,1,0),c(-1,0,0,1),c(0,-1,1,0),c(0,-1,0,1),c(0,0,-1,1))

N <- 1000;
FGH(n,p,mu,sigma,N,C2 = C2)

## End(Not run)

DataXujing/LN0SCIs documentation built on May 24, 2019, 9:52 a.m.