calpredict(chemCal)R Documentation

Estimate measurement results including confidence intervals

Description

This function generates estimates for x values from y values, including a confidence interval for the x values. The formulas in this function used for prediction of concentrations from (replicate) measurements are taken from the "Handbook of Chemometrics and Qualimetrics Part A" by D. L. Massart, Vandeginste, B. G. M., Buydens, L. M. C., De Jong, S., Lewi, P. J. and Smeyers-Verbeke, J, Elsevier, Amsterdam, 1997 and from the EURACHEM/CITAC report on "Quantifying uncertainty in analytical measurement", 2000, pp. 111f.

Usage

  calpredict(yobs,xi,yi,xunit="",level=0.95,intercept=FALSE,syobs=FALSE)

Arguments

yobs A numeric vector containing the observed data.
xi A vector of x values of the calibration.
yi A vector of y values of the calibration.
xunit The unit of the given values on the x axis as a character string.
level The desired confidence level for the confidence interval of the estimates. Defaults to 0.95.
intercept Logical value determining if an intercept is to be fitted or not. Default is FALSE.
syobs If TRUE, a standard deviation for the given y values is calculated, and the resulting confidence interval will include this variability (not validated yet). If FALSE (default), this standard deviation is not included in the confidence interval. If a numeric value is given, it is used for the standard deviation of "real samples", in addition to the standard deviation of the y values in the calibration (also not validated yet).

Value

A list containing the estimate, its standard deviation and its confidence interval.

Author(s)

Johannes Ranke jranke@uni-bremen.de http://www.uft.uni-bremen.de/chemie/ranke

Examples

data(pahCalibration)
attach(pahCalibration)
y <- c(51.2,51.4,51.1,51.8)
estimate <- calpredict(y,conc,acenaphthene,xunit="mg/L")
[Package Contents]