snsp2mp {SenSpe}

Two-biomarker paired comparison in specificity (or sensitivity) at a controlled sensitivity (or specificity) level

Description

Point estimation and exact bootstrap-based inference

Usage

snsp2mp(mk, n1, s0, covp=0.95, fixsens=TRUE, lbmdis=TRUE)

Arguments

mk

Each of two rows corresponds to a biomarker, cases followed by controls.

n1

case size.

s0

controlled level of sensitivity or specificity.

covp

norminal level of confidence intervals.

fixsens

fixing sensitivity if True, and specificity otherwise.

lbmdis

larger value of a biomarker is more associated with cases if True, and controls otherwise.

Value

diff

diff[1]: difference of empirical point estimates; hss[2]: difference of oscillating bias-corrected estimates.

btmn

bootstrap mean of the empirical difference.

btva

exact bootstrap variance estimate for diff[1].

btdist

exact bootstrap probability mass function at (-n0:n0)/n0 with n0 being the size of controls if sensitivity is controlled, or at (-n1:n1)/n1 otherwise.

wald_ci

wald_ci[1,]: Wald confidence interval using diff[1]; wald_ci[2,]: Wald confidence interval using diff[2].

pct_ci

percentile confidence interval.

scr_ci

scr_ci[1,]: score confidence interval using diff[1]; scr_ci[2,]: score confidence interval using diff[2].

zq_ci

extension of the BTII in Zhou and Qin (2005, Statistics in Medicine 24, pp 465–477).

Author(s)

Yijian Huang

References

Huang, Y., Parakati, I., Patil, D. H.,and Sanda, M. G. (2023). Interval estimation for operating characteristic of continuous biomarkers with controlled sensitivity or specificity, Statistica Sinica 33, 193–214.

Examples

## simulate paired biomarkers X and Y, with correlation 0.5, 100 cases and 100 controls
n1 <- 100
n0 <- 100
rho <- 0.5
set.seed(1234)
mkx <- rnorm(n1+n0,0,1)
mky <- rho*mkx + sqrt(1-rho^2)*rnorm(n1+n0,0,1)
mkx <- mkx + c(rep(2,n1),rep(0,n0))
mky <- mky + c(rep(1,n1),rep(0,n0))
mk <- rbind(mkx,mky)

## compare specificity at controlled 0.95 sensitivity
est <- snsp2mp(mk, 100, 0.95)