diff options
author | ranke <ranke@5fad18fb-23f0-0310-ab10-e59a3bee62b4> | 2005-02-15 10:14:01 +0000 |
---|---|---|
committer | ranke <ranke@5fad18fb-23f0-0310-ab10-e59a3bee62b4> | 2005-02-15 10:14:01 +0000 |
commit | a94bd86465fe191102a2bf85a3631c83cd10db0a (patch) | |
tree | 489388209c63519def400a55c2ecffe29ba1e56c /man/calpredict.Rd |
First import, for archiving purposes.
git-svn-id: http://kriemhild.uft.uni-bremen.de/svn/chemCal@1 5fad18fb-23f0-0310-ab10-e59a3bee62b4
Diffstat (limited to 'man/calpredict.Rd')
-rw-r--r-- | man/calpredict.Rd | 64 |
1 files changed, 64 insertions, 0 deletions
diff --git a/man/calpredict.Rd b/man/calpredict.Rd new file mode 100644 index 0000000..a91d018 --- /dev/null +++ b/man/calpredict.Rd @@ -0,0 +1,64 @@ +\name{calpredict} +\alias{calpredict} +\title{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{ + \item{yobs}{ + A numeric vector containing the observed data. + } + \item{xi}{ + A vector of x values of the calibration. + } + \item{yi}{ + A vector of y values of the calibration. + } + \item{xunit}{ + The unit of the given values on the x axis as a character string. + } + \item{level}{ + The desired confidence level for the confidence interval of the + estimates. Defaults to 0.95. + } + \item{intercept}{ + Logical value determining if an intercept is to be fitted or not. + Default is FALSE. + } + \item{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. +} +\examples{ +data(pahCalibration) +attach(pahCalibration) +y <- c(51.2,51.4,51.1,51.8) +estimate <- calpredict(y,conc,acenaphthene,xunit="mg/L") +} +\author{ + Johannes Ranke + \email{jranke@uni-bremen.de} + \url{http://www.uft.uni-bremen.de/chemie/ranke} +} +\keyword{regression} |