From 3b3d6dfc88c4b8b6475147a3afb5258a5fc82fa5 Mon Sep 17 00:00:00 2001
From: ranke <ranke@5fad18fb-23f0-0310-ab10-e59a3bee62b4>
Date: Tue, 23 May 2006 15:24:58 +0000
Subject: First version published on my website.

git-svn-id: http://kriemhild.uft.uni-bremen.de/svn/chemCal@14 5fad18fb-23f0-0310-ab10-e59a3bee62b4
---
 man/calplot.lm.Rd      | 26 +++++++++++++++-----------
 man/chemCal-package.Rd |  8 ++++----
 man/lod.Rd             |  3 +++
 man/massart97ex3.Rd    | 14 +++++++++-----
 4 files changed, 31 insertions(+), 20 deletions(-)

(limited to 'man')

diff --git a/man/calplot.lm.Rd b/man/calplot.lm.Rd
index 734933d..de63022 100644
--- a/man/calplot.lm.Rd
+++ b/man/calplot.lm.Rd
@@ -8,8 +8,9 @@
   as prediction and confidence bands. 
 }
 \usage{
-  calplot(object, xlim = "auto", ylim = "auto", 
-  xlab = "Concentration", ylab = "Response", alpha=0.05)
+  calplot(object, xlim = c("auto","auto"), ylim = c("auto","auto"),
+  xlab = "Concentration", ylab = "Response", alpha=0.05, 
+  varfunc = NULL)
 }
 \arguments{
   \item{object}{
@@ -32,22 +33,25 @@
   \item{alpha}{
     The error tolerance level for the confidence and prediction bands.
   }
+  \item{varfunc}{
+    The variance function for generating the weights in the model.
+    Currently, this argument is ignored (see note below).
+  }
 }
 \value{
   A plot of the calibration data, of your fitted model as well as lines showing
   the confidence limits as well as the prediction limits.
 } 
+\note{
+  Prediction bands for models from weighted linear regression require weights
+  for the data, for which responses should be predicted. Prediction intervals
+  for weighted models are not currently supported by the internally used
+  function \code{\link{predict.lm}}, therefore, \code{calplot} refuses to work
+  for such models.
+}
 \examples{
-# Example of a Calibration plot for a weighted regression
-source("/home/ranke/tmp/r-base-2.3.0/src/library/stats/R/lm.R")
 data(massart97ex3)
-attach(massart97ex3)
-yx <- split(y,factor(x))
-s <- round(sapply(yx,sd),digits=2)
-w <- round(1/(s^2),digits=3)
-weights <- w[factor(x)]
-m <- lm(y ~ x,w=10 * weights)
-calplot(m)
+m <- lm(y ~ x, data=massart97ex3)
 calplot(m)
 }
 \author{
diff --git a/man/chemCal-package.Rd b/man/chemCal-package.Rd
index a11744e..4456150 100644
--- a/man/chemCal-package.Rd
+++ b/man/chemCal-package.Rd
@@ -2,16 +2,16 @@
 \alias{chemCal-package}
 \docType{package}
 \title{
-Calibration functions for analytical chemistry
+  Calibration functions for analytical chemistry
 }
 \description{
-See \url{../DESCRIPTION}
+  See \url{../DESCRIPTION}
 }
 \details{
-There is a package vignette located in \url{../doc/chemCal.pdf}.
+  There is a package vignette located in \url{../doc/chemCal.pdf}.
 }
 \author{
-Author and Maintainer: Johannes Ranke <jranke@uni-bremen.de>
+  Author and Maintainer: Johannes Ranke <jranke@uni-bremen.de>
 }
 \keyword{manip}
 }
diff --git a/man/lod.Rd b/man/lod.Rd
index fa8c8ad..c2fe4e9 100644
--- a/man/lod.Rd
+++ b/man/lod.Rd
@@ -43,6 +43,9 @@
   - The estimation of the LOD in terms of the analyte amount/concentration
     xD from the LOD in the signal domain SD is done by simply inverting the
     calibration function (i.e. assuming a known calibration function).
+  - The calculation of a LOD from weighted calibration models requires
+    a weights argument for the internally used \code{\link{predict.lm}}
+    function, which is currently not supported in R.
 }
 \references{
   Massart, L.M, Vandenginste, B.G.M., Buydens, L.M.C., De Jong, S., Lewi, P.J., 
diff --git a/man/massart97ex3.Rd b/man/massart97ex3.Rd
index 2618709..eb00e79 100644
--- a/man/massart97ex3.Rd
+++ b/man/massart97ex3.Rd
@@ -23,18 +23,22 @@ m <- lm(y ~ x,w=weights)
 inverse.predict(m, 15, ws = 1.67)
 inverse.predict(m, 90, ws = 0.145)
 
-calplot(m)
+# Some of the following examples are commented out, because the require
+# prediction intervals from predict.lm for weighted models, which is not
+# available in R at the moment.
+
+#calplot(m)
 
 m0 <- lm(y ~ x)
 lod(m0)
-lod(m)
+#lod(m)
 
 # Now we want to take advantage of the lower weights at lower y values
-m2 <- lm(y ~ x, w = 1/y)
+#m2 <- lm(y ~ x, w = 1/y)
 # To get a reasonable weight for the lod, we need to estimate it and predict 
 # a y value for it
-yhat.lod <- predict(m,data.frame(x = lod(m2)))
-lod(m2,w=1/yhat.lod,k=3)
+#yhat.lod <- predict(m,data.frame(x = lod(m2)))
+#lod(m2,w=1/yhat.lod,k=3)
 }
 \source{
   Massart, L.M, Vandenginste, B.G.M., Buydens, L.M.C., De Jong, S., Lewi, P.J., 
-- 
cgit v1.2.1