calfun {CVcalibration} | R Documentation |
Estimating the calibration equation “y=a+b*x” with error-in observations assuming that the coefficients of the variation of the measurements are constants.
calfun(x, y, CVx, CVy, lambda0)
x |
The observed $x$ values |
y |
The observed $y$ values |
CVx |
The underlying coefficient of variation of measurement $x$ |
CVy |
The underlying coefficient of variation of measurement $y$ |
lambda0 |
The ratio, $CV_y^2/CV_x^2$ |
result |
The estimated regression coefficients, standard error and confidence intervals based on (1) CVx only; (2) CVy only; (3) both CVx and CVy; and (4) the ratio of CVy^2/CVx^2. |
Lu Tian, He Qi
n=100 sigma0=10 beta0=5 beta1=1.2 CVx=0.15 CVy=0.07 lambda0=CVy^2/CVx^2 x0=runif(n, 20, 200) y0=beta0+beta1*x0+rnorm(n)*sigma0 x=x0+x0*CVx*rnorm(n) y=y0+y0*CVy*rnorm(n) fit=calfun(x, y, CVx, CVy, lambda0) fit