[Dplr-commits] r1114 - in pkg/dplR: . R man

[noreply at r-forge.r-project.org]

Author: andybunn
Date: 2018-06-14 13:32:01 +0200 (Thu, 14 Jun 2018)
New Revision: 1114

Modified:
   pkg/dplR/ChangeLog
   pkg/dplR/R/rwl.stats.R
   pkg/dplR/man/rwl.stats.Rd
Log:
Removing sens1 and sens2 from rwl.stats.

Modified: pkg/dplR/ChangeLog
===================================================================
--- pkg/dplR/ChangeLog	2018-06-14 11:16:00 UTC (rev 1113)
+++ pkg/dplR/ChangeLog	2018-06-14 11:32:01 UTC (rev 1114)
@@ -1,10 +1,18 @@
 * CHANGES IN dplR VERSION 1.6.9
 
+File: rwl.stats.R and .Rd
+----------------
+
+- Removing sens1 and sens2 from this function. At world dendro I'm still seeing these being used by dplR users. It's a terrible stat. I'm taking it out.  Help file ammended.
+
 File: detrend.series.R and .Rd
 ----------------
 
 - The function will now return the curves used for detrnding the series if return.info is TRUE.  Help file ammended.
 
+- Added the Hughershoff curve as an method for detrending. It's done along the lines of ModNegExp with straight line if the nls call fais. 
+
+
 File: detrend.R
 ----------------
 

Modified: pkg/dplR/R/rwl.stats.R
===================================================================
--- pkg/dplR/R/rwl.stats.R	2018-06-14 11:16:00 UTC (rev 1113)
+++ pkg/dplR/R/rwl.stats.R	2018-06-14 11:32:01 UTC (rev 1114)
@@ -23,8 +23,8 @@
     series.stats$median <- colMedians(rwl2, na.rm=TRUE)
     series.stats$stdev <- colSds(rwl2, na.rm=TRUE)
     series.stats$skew <- apply(rwl2, 2, skew)
-    series.stats$sens1 <- apply(rwl2, 2, sens1)
-    series.stats$sens2 <- apply(rwl2, 2, sens2)
+    #series.stats$sens1 <- apply(rwl2, 2, sens1)
+    #series.stats$sens2 <- apply(rwl2, 2, sens2)
     series.stats$gini <- apply(rwl2, 2, gini.coef)
     series.stats$ar1 <- apply(rwl2, 2, acf1)
     seq.temp <- -seq_len(4)

Modified: pkg/dplR/man/rwl.stats.Rd
===================================================================
--- pkg/dplR/man/rwl.stats.Rd	2018-06-14 11:16:00 UTC (rev 1113)
+++ pkg/dplR/man/rwl.stats.Rd	2018-06-14 11:32:01 UTC (rev 1114)
@@ -42,12 +42,10 @@
   well as the length of the series (\code{"first"}, \code{"last"},
   \code{"year"}).  The mean, median, standard deviation are given
   (\code{"mean"}, \code{"median"}, \code{"stdev"}) as are the skewness,
-  two measures of sensitivity, the Gini coefficient, and first order
-  autocorrelation (\code{"skew"}, \code{"\link{sens1}"},
-  \code{"\link{sens2}"}, \code{"\link{gini.coef}"}, \code{"ar1"}).
+  the Gini coefficient, and first order
+  autocorrelation (\code{"skew"}, \code{"\link{gini.coef}"}, \code{"ar1"}).
   
-  Note that mean sensitivity is not a robust statistic that should rarely,
-  if ever, be used (Bunn et al. 2013).
+  Note that prior to version 1.6.8, two measures of sensitivity were also included. However mean sensitivity is not a robust statistic that should rarely, if ever, be used (Bunn et al. 2013). Those sensitivity functions (\code{"\link{sens1}"} and \code{"\link{sens2}"}) are still available for continuity. Users should consider the coef of variation in lieu of mean sensitivity.
 }
 \references{