[Dplr-commits] r719 - / pkg/dplR pkg/dplR/man
noreply at r-forge.r-project.org
noreply at r-forge.r-project.org
Sun Jan 12 02:31:43 CET 2014
Author: andybunn
Date: 2014-01-12 02:31:43 +0100 (Sun, 12 Jan 2014)
New Revision: 719
Modified:
/
pkg/dplR/ChangeLog
pkg/dplR/man/tbrm.Rd
Log:
fixed tbrm and modified its example slightly.
Property changes on:
___________________________________________________________________
Modified: svn:ignore
- .Rproj.user
.Rhistory
.RData
+ .Rproj.user
.Rhistory
.RData
dplr.Rproj
Modified: pkg/dplR/ChangeLog
===================================================================
--- pkg/dplR/ChangeLog 2014-01-11 06:29:53 UTC (rev 718)
+++ pkg/dplR/ChangeLog 2014-01-12 01:31:43 UTC (rev 719)
@@ -1,13 +1,47 @@
* CHANGES IN dplR VERSION 1.5.8
+File: tbrm.Rd
+---------------
+- Improved (slightly) the trbm() examples.
+
+
File: tbrm.R
---------------
- Changed tbrm call to .C so that it doesn't
break CRAN's anti-social policy. Email from
- BDR details error but R CMD check --as-cran dplR_1.5.8.tar.gz
- still fails.
+ BDR details error.
+ Begin quote from Ripley:
+ The error is in your function tbrm:
+
+ > tbrm
+ function (x, C = 9)
+ {
+ y <- as.double(x[!is.na(x)])
+ n <- as.integer(length(y))
+ stopifnot(!is.na(n))
+ .C(dplR.tbrm,
+ y, n, as.double(C), result = NaN, NAOK = TRUE, DUP = FALSE)$result
+ }
+
+ That .C call changes its arguments: see the warnings in ?.C . In this
+ case it changes the binding of NaN.
+
+ You could remove DUP = FALSE (these days it is not helping) or use
+
+ > tbrm
+ function (x, C = 9)
+ {
+ y <- as.double(x[!is.na(x)])
+ n <- as.integer(length(y))
+ stopifnot(!is.na(n))
+ .C(dplR.tbrm, y, n, as.double(C), result = double(1L), NAOK = TRUE,
+ DUP = FALSE)$result
+ }
+ End quote from Ripley
+ This change was made and R CMD check --as-cran passes fine.
+
File: DESCRIPTION
- Added Jacob Cecile as a contributor
Modified: pkg/dplR/man/tbrm.Rd
===================================================================
--- pkg/dplR/man/tbrm.Rd 2014-01-11 06:29:53 UTC (rev 718)
+++ pkg/dplR/man/tbrm.Rd 2014-01-12 01:31:43 UTC (rev 719)
@@ -1,64 +1,68 @@
-\name{tbrm}
-\alias{tbrm}
-\title{ Calculate Tukey's Biweight Robust Mean }
-\description{
- This calculates a robust average that is unaffected by outliers.
-}
-\usage{
-tbrm(x, C = 9)
-}
-\arguments{
- \item{x}{ a \code{numeric} vector }
- \item{C}{ a constant. \code{\var{C}} is preassigned a value of 9
- according to the Cook reference below but other values are
- possible. }
-}
-\details{
- This is a one step computation that follows the Affy whitepaper below,
- see page 22. This function is called by \code{\link{chron}} to
- calculate a robust mean. \code{\var{C}} determines the point at which
- outliers are given a weight of 0 and therefore do not contribute to
- the calculation of the mean. \code{\var{C} = 9} sets values roughly
- +/-6 standard deviations to 0. \code{\var{C} = 6} is also used in
- tree-ring chronology development. Cook and Kairiukstis (1990) have
- further details.
-
- An exact summation algorithm (Shewchuk 1997) is used. When some
- assumptions about the rounding of floating point numbers and
- conservative compiler optimizations hold, summation error is
- completely avoided. Whether the assumptions hold depends on the
- platform, i.e. compiler and \acronym{CPU}.
-}
-\value{
- A \code{numeric} mean.
-}
-\references{
-
- Statistical Algorithms Description Document, 2002, Affymetrix.
-
- Cook, E. R. and Kairiukstis, L. A. (1990) \emph{Methods of
- Dendrochronology: Applications in the Environmental Sciences}.
- Springer. \acronym{ISBN-13}: 978-0-7923-0586-6.
-
- Mosteller, F. and Tukey, J. W. (1977) \emph{Data Analysis and
- Regression: a second course in statistics}. Addison-Wesley.
- \acronym{ISBN-13}: 978-0-201-04854-4.
-
- Shewchuk, J. R. (1997) Adaptive Precision Floating-Point Arithmetic
- and Fast Robust Geometric Predicates. \emph{Discrete and
- Computational Geometry}, 18(3):305\enc{–}{--}363. Springer.
-
-}
-\author{ Mikko Korpela }
-\seealso{ \code{\link{chron}} }
-\examples{tbrm(rnorm(100))
-
-## Compare
-data(co021)
-co021.rwi <- detrend(co021, method = "ModNegExp")
-crn1 <- apply(co021.rwi, 1, tbrm)
-crn2 <- chron(co021.rwi)
-cor(crn1, crn2[, 1])
-}
-\keyword{ robust }
-\keyword{ univar }
+\name{tbrm}
+\alias{tbrm}
+\title{ Calculate Tukey's Biweight Robust Mean }
+\description{
+ This calculates a robust average that is unaffected by outliers.
+}
+\usage{
+tbrm(x, C = 9)
+}
+\arguments{
+ \item{x}{ a \code{numeric} vector }
+ \item{C}{ a constant. \code{\var{C}} is preassigned a value of 9
+ according to the Cook reference below but other values are
+ possible. }
+}
+\details{
+ This is a one step computation that follows the Affy whitepaper below,
+ see page 22. This function is called by \code{\link{chron}} to
+ calculate a robust mean. \code{\var{C}} determines the point at which
+ outliers are given a weight of 0 and therefore do not contribute to
+ the calculation of the mean. \code{\var{C} = 9} sets values roughly
+ +/-6 standard deviations to 0. \code{\var{C} = 6} is also used in
+ tree-ring chronology development. Cook and Kairiukstis (1990) have
+ further details.
+
+ An exact summation algorithm (Shewchuk 1997) is used. When some
+ assumptions about the rounding of floating point numbers and
+ conservative compiler optimizations hold, summation error is
+ completely avoided. Whether the assumptions hold depends on the
+ platform, i.e. compiler and \acronym{CPU}.
+}
+\value{
+ A \code{numeric} mean.
+}
+\references{
+
+ Statistical Algorithms Description Document, 2002, Affymetrix.
+
+ Cook, E. R. and Kairiukstis, L. A. (1990) \emph{Methods of
+ Dendrochronology: Applications in the Environmental Sciences}.
+ Springer. \acronym{ISBN-13}: 978-0-7923-0586-6.
+
+ Mosteller, F. and Tukey, J. W. (1977) \emph{Data Analysis and
+ Regression: a second course in statistics}. Addison-Wesley.
+ \acronym{ISBN-13}: 978-0-201-04854-4.
+
+ Shewchuk, J. R. (1997) Adaptive Precision Floating-Point Arithmetic
+ and Fast Robust Geometric Predicates. \emph{Discrete and
+ Computational Geometry}, 18(3):305\enc{–}{--}363. Springer.
+
+}
+\author{ Mikko Korpela }
+\seealso{ \code{\link{chron}} }
+\examples{
+
+foo <- rnorm(100)
+tbrm(foo)
+mean(foo)
+
+## Compare
+data(co021)
+co021.rwi <- detrend(co021, method = "ModNegExp")
+crn1 <- apply(co021.rwi, 1, tbrm)
+crn2 <- chron(co021.rwi)
+cor(crn1, crn2[, 1])
+}
+\keyword{ robust }
+\keyword{ univar }
More information about the Dplr-commits
mailing list