DiagTest3Grp (version 1.6)

Youden3Grp.Variance.Bootstrap: Obtain bootstrap variance and confidence itnerval on the extended Youden index and optimal cut-off points

Description

Given the diagnostic test measurements x, y, z in the three ordinal groups $D^-$, $D^0$, $D^+$ separately, the function repeatedly draws a bootstrap sample each of x, y, z to estimate the extended Youden index for three ordinal groups and the associated optimal cut point and obtain the final bootstrap average estimate on the Youden index and optimal cut-off point and the confidence interval.

Usage

Youden3Grp.Variance.Bootstrap(x, y, z, method = "Normal", seed.seq =NULL,
randomStart.N = 5, NBOOT=10,alpha=0.05,t.minus.start = NULL,
t.plus.start = NULL, ...)

Arguments

x

A numeric vector, a diagnostic test's measurements for subjects in D- (usually a healthy group).

y

A numeric vector, a diagnostic test's measurements for subjects in D0 (usually a mildly diseased group).

z

A numeric vector, a diagnostic test's measurements for subjects in D+ (usually a severly diseased group).

method

A character argument. Specify a method to be used for estimating the extended Youden index Choices include ‘Normal’---Estimate the extended Youden index under the assumption of normal distributions of a diagnostic test in the three groups. ‘TN’---Transformed normal. Implement Box-cox transformation to approximate normality and then estimate the extended Youden index under normality. ‘EMP’---Estimate the extended Youden index by using empirical cumulative density function. ‘KS’---Estimate the extended Youden index by using Kernel density estimation with a normal reference rule for bandwidth selection.

‘KS-SJ’---Estimate the extended Youden index by using Kernel density estimation with the Sheather-Jones Plug-in method for bandwidth selection.

seed.seq

A numeric vector. Users can specify a sequence of random seeds for bootstrap sampling of x. Boostrapping of y and z will use seed0+1 and seed0+2. Default will be the sequence of 1:10.

randomStart.N

A numeric value. An argument need to be specified when the method “EMP”, “KS” or “KS-SJ” is used to estimate the optimal cut-point and the extended Youden index J. Default, randomStart.N=1. See Youden3Grp.

NBOOT

A numeric value. Total number of bootstrapping, default=10.

t.minus.start

The starting points of the lower optimal cut-point (t-) which separate the D- and D0 group. Default t.minus.start=NULL will randomly generate starting value. See Youden3Grp.

t.plus.start

A numeric value. The starting points of the upper optimal cut-point (t+) which separate the D0 and D+ group. Default t.plus.start=NULL will randomly generate starting value. See Youden3Grp.

alpha

A numeric value. the significance level, will provide the basic quantile confidence interval (alpha/2*100%,1-alpha/2*100%).

...

Other arguments to be passed to the R function Youden3Grp.

Value

Return a list, including the following components

var.youden

The normal-method based variance on the optimal Youden index.

var.t.minus

A numeric value.The variance on the lower optimal cut-point t.minus.

var.t.plus

A numeric value.The variance on the upper optimal cut-point t.plus.

var.youden.z

A numeric value. The variance on the Fisher's Z transformed optimal Youden index.

youden.CI

A named numeric of length 2. CI for the estimate of youden with the lower and the upper CI limit (under the names alpha/2*100% and (1-alpha/2)*100%).

t.minus.CI

A named numeric of length 2. CI for the estimate of t.minus (the lower optimal cut-point) with the lower and the upper CI limit (under the names alpha/2*100% and (1-alpha/2)*100%) CI

t.plus.CI

A named numeric of length 2. CI for the estimate of t.plus (the upper optimal cut-point) with the lower and the upper CI limit (under the names alpha/2*100% and (1-alpha/2)*100%).

youden.z.CI

A named numeric of length 2. CI for the estimate of Fisher-Z transformed youden with the lower and the upper CI limit (under the names alpha/2*100% and (1-alpha/2)*100%).

Details

This function is carried to get bootstrap estimates of the extended Youden index and associated cut-points to provide confidence interval. See details in Youden3Grp

References

Luo, J and Xiong, C. (2012) Youden Index and Associated Optimal Cut-point for Three Ordinal Groups. Communications In Statistics-Simulation and Computation (in press).

See Also

Youden3Grp BW.ref

Examples

x <- rnorm(50,6,1)

y <- rnorm(60,8,1.2)

z <- rnorm(40,10,1.4)

temp.res <- Youden3Grp.Variance.Bootstrap(x=x, y=y, z=z,
method="Normal",seed.seq=1:10,randomStart.N=3,NBOOT=10)