From fd82f17fb65a0abe86970e92e2468ef10ec7209d Mon Sep 17 00:00:00 2001 From: Walmes Zeviani <walmeszeviani@gmail.com> Date: Sun, 13 Sep 2015 23:04:08 -0300 Subject: [PATCH] Documents the defoliation dataset. --- R/legTools.R | 124 +++++++++++++++++++++++++++++++++++++-------- man/defoliation.Rd | 83 ++++++++++++++++++++++++++++++ 2 files changed, 187 insertions(+), 20 deletions(-) create mode 100644 man/defoliation.Rd diff --git a/R/legTools.R b/R/legTools.R index bd4334f..9cbf85e 100644 --- a/R/legTools.R +++ b/R/legTools.R @@ -12,11 +12,12 @@ NULL #' @title Feeding type in pig weight gain #' #' @description This is an artifial dataset corresponding a experiment -#' to study the effect of feeding type (factor with 4 categorical -#' nominal levels) in pig weight gain. The experiment was a randomized -#' complete design with five experimental units per treatment level. The -#' experimental unit was a pig. The response measured was weight gain -#' from the beggining to the end of the experiment. +#' to study the effect of feeding type (factor with 4 categorical +#' nominal levels) in pig weight gain. The experiment was a +#' randomized complete design with five experimental units per +#' treatment level. The experimental unit was a pig. The response +#' measured was weight gain from the beggining to the end of the +#' experiment. #' #' \itemize{ #' \item \code{ft} feeding type, a categorical factor with 4 @@ -32,8 +33,8 @@ NULL #' #' @format a \code{data.frame} with 20 records and 2 variables. #' -#' @source Frederico, P. (2009). Curso de Estatística Experimental -#' (15th ed.). Piracicaba, São Paulo: FEALQ. (page 62) +#' @source Frederico, P. (2009). Curso de Estatística Experimental (15th +#' ed.). Piracicaba, São Paulo: FEALQ. (page 62) #' #' @examples #' @@ -51,9 +52,9 @@ NULL #' @title Potato variety competition experiment #' #' @description These data are from an experiment done by the engineer -#' Oscar A. Garay at Balcare, Argentina. The experiment was done in a -#' randomized complete block design with 4 blocks. Potato yield (t/ha) -#' was recorded in each experimental unit. +#' Oscar A. Garay at Balcare, Argentina. The experiment was done in +#' a randomized complete block design with 4 blocks. Potato yield +#' (t/ha) was recorded in each experimental unit. #' #' \itemize{ #' \item \code{block} a categorical unordered factor with 4 levels. @@ -70,8 +71,8 @@ NULL #' #' @format a \code{data.frame} with 32 records and 3 variables. #' -#' @source Frederico, P. (2009). Curso de Estatística Experimental -#' (15th ed.). Piracicaba, São Paulo: FEALQ. (page 76) +#' @source Frederico, P. (2009). Curso de Estatística Experimental (15th +#' ed.). Piracicaba, São Paulo: FEALQ. (page 76) #' #' @examples #' require(lattice) @@ -89,12 +90,12 @@ NULL #' @title Plowing level on corn yield #' #' @description These data are from an experiment done by the engineer -#' Duvilio Ometto to study the effect of plowing level on corn yield. It -#' was used 2 levels of plowing: normal (or superficial) and deep. The -#' experiment was done in a randomized complete block design with 6 -#' blocks. Corn yield (t/ha) was recorded in each experimental unit -#' but in this experiment there was 2 experimental units for each factor -#' level in each block. +#' Duvilio Ometto to study the effect of plowing level on corn +#' yield. It was used 2 levels of plowing: normal (or superficial) +#' and deep. The experiment was done in a randomized complete block +#' design with 6 blocks. Corn yield (t/ha) was recorded in each +#' experimental unit but in this experiment there was 2 experimental +#' units for each factor level in each block. #' #' \itemize{ #' \item \code{block} a categorical unordered factor with 6 levels. @@ -110,8 +111,8 @@ NULL #' #' @format a \code{data.frame} with 24 records and 3 variables. #' -#' @source Frederico, P. (2009). Curso de Estatística Experimental -#' (15th ed.). Piracicaba, São Paulo: FEALQ. (page 91) +#' @source Frederico, P. (2009). Curso de Estatística Experimental (15th +#' ed.). Piracicaba, São Paulo: FEALQ. (page 91) #' #' @examples #' require(lattice) @@ -122,3 +123,86 @@ NULL #' xlab="Plowing level") #' NULL + +#' @name defoliation +#' +#' @title Bolls in cotton as function of artifitial defoliation +#' +#' @description This dataset contais the result of a real experiment to +#' evaluate the effect of artifitial defoliation in combination with +#' phenological stage of occurence on the production of cotton +#' represented by the number of bolls produced at the end of the +#' crop cycle. The experiment is a \eqn{5\times 5} factorial with 5 +#' replications casualized at random to the experimental units (a +#' randomized complete design). The experimental unit was a pot with +#' 2 plants. An interesting fact about this data is that the +#' response is a count variable that shows underdispersion (sample +#' variance less than the sample mean). +#' +#' \itemize{ +#' \item \code{phenol} a categorical ordered factor with 5 levels +#' that represent the phenological stages of the cotton plant in +#' which defoliation was applied. +#' \item \code{defol} a numeric factor with 5 levels that represents the +#' artifical level of defoliation (percent in leaf area removed with +#' scissors) applied for all leaves in the plant. +#' \item \code{rept} index for each experimenal unit in each treatment cell. +#' \item \code{bolls} the number of bolls produced (count variable) +#' evaluated at harvest. +#' } +#' +#' @details The experiment was done in a greenhouse at Universidade +#' Federal da Grande Dourados. Visit +#' \itemize{ +#' \item 1) \code{http://www.cabdirect.org/abstracts/20123299470.html} +#' \item 2) \code{http://leg.ufpr.br/doku.php/publications:papercompanions:zeviani-jas2014} +#' } +#' 1 for an article discussing the effect of defoliation on cotton yield and +#' visit 2 for an article that evaluate a count regression model able to +#' deal with the underdispersion. See the references section also. +#' +#' @docType data +#' +#' @keywords datasets +#' +#' @usage data(defoliation) +#' +#' @format a \code{data.frame} with 125 records and 4 variables. +#' +#' @references Silva, A. M., Degrande, P. E., Suekane, R., Fernandes, +#' M. G., & Zeviani, W. M. (2012). Impacto de diferentes níveis de +#' desfolha artificial nos estádios fenológicos do +#' algodoeiro. Revista de Ciências Agrárias, 35(1), 163–172. +#' +#' Zeviani, W. M., Ribeiro, P. J., Bonat, W. H., Shimakura, S. E., & +#' Muniz, J. A. (2014). The Gamma-count distribution in the analysis +#' of experimental underdispersed data. Journal of Applied +#' Statistics, 41(12), +#' 1–11. http://doi.org/10.1080/02664763.2014.922168 +#' +#' @examples +#' +#' library(lattice) +#' library(latticeExtra) +#' +#' ## x11(width=7, height=2.8) +#' xyplot(bolls~defol|phenol, data=defoliation, +#' layout=c(NA, 1), type=c("p", "smooth"), +#' xlab="Artificial defoliation level", +#' ylab="Number of bolls produced", +#' xlim=extendrange(c(0:1), f=0.15), jitter.x=TRUE) +#' +#' ## Sample mean and variance in each treatment cell. +#' mv <- aggregate(bolls~phenol+defol, data=defoliation, +#' FUN=function(x) c(mean=mean(x), var=var(x))) +#' str(mv) +#' +#' xlim <- ylim <- extendrange(c(mv$bolls), f=0.05) +#' +#' ## Evidence in favor of the underdispersion. +#' xyplot(bolls[,"var"]~bolls[,"mean"], data=mv, +#' aspect="iso", xlim=xlim, ylim=ylim, +#' ylab="Sample variance", xlab="Sample mean")+ +#' layer(panel.abline(a=0, b=1, lty=2)) +#' +NULL diff --git a/man/defoliation.Rd b/man/defoliation.Rd new file mode 100644 index 0000000..d6ea686 --- /dev/null +++ b/man/defoliation.Rd @@ -0,0 +1,83 @@ +% Generated by roxygen2 (4.1.1): do not edit by hand +% Please edit documentation in R/legTools.R +\docType{data} +\name{defoliation} +\alias{defoliation} +\title{Bolls in cotton as function of artifitial defoliation} +\format{a \code{data.frame} with 125 records and 4 variables.} +\usage{ +data(defoliation) +} +\description{ +This dataset contais the result of a real experiment to + evaluate the effect of artifitial defoliation in combination with + phenological stage of occurence on the production of cotton + represented by the number of bolls produced at the end of the + crop cycle. The experiment is a \eqn{5\times 5} factorial with 5 + replications casualized at random to the experimental units (a + randomized complete design). The experimental unit was a pot with + 2 plants. An interesting fact about this data is that the + response is a count variable that shows underdispersion (sample + variance less than the sample mean). + +\itemize{ +\item \code{phenol} a categorical ordered factor with 5 levels + that represent the phenological stages of the cotton plant in + which defoliation was applied. +\item \code{defol} a numeric factor with 5 levels that represents the + artifical level of defoliation (percent in leaf area removed with + scissors) applied for all leaves in the plant. +\item \code{rept} index for each experimenal unit in each treatment cell. +\item \code{bolls} the number of bolls produced (count variable) + evaluated at harvest. +} +} +\details{ +The experiment was done in a greenhouse at Universidade + Federal da Grande Dourados. Visit +\itemize{ +\item 1) \code{http://www.cabdirect.org/abstracts/20123299470.html} +\item 2) \code{http://leg.ufpr.br/doku.php/publications:papercompanions:zeviani-jas2014} +} +1 for an article discussing the effect of defoliation on cotton yield and + visit 2 for an article that evaluate a count regression model able to + deal with the underdispersion. See the references section also. +} +\examples{ +library(lattice) +library(latticeExtra) + +## x11(width=7, height=2.8) +xyplot(bolls~defol|phenol, data=defoliation, + layout=c(NA, 1), type=c("p", "smooth"), + xlab="Artificial defoliation level", + ylab="Number of bolls produced", + xlim=extendrange(c(0:1), f=0.15), jitter.x=TRUE) + +## Sample mean and variance in each treatment cell. +mv <- aggregate(bolls~phenol+defol, data=defoliation, + FUN=function(x) c(mean=mean(x), var=var(x))) +str(mv) + +xlim <- ylim <- extendrange(c(mv$bolls), f=0.05) + +## Evidence in favor of the underdispersion. +xyplot(bolls[,"var"]~bolls[,"mean"], data=mv, + aspect="iso", xlim=xlim, ylim=ylim, + ylab="Sample variance", xlab="Sample mean")+ + layer(panel.abline(a=0, b=1, lty=2)) +} +\references{ +Silva, A. M., Degrande, P. E., Suekane, R., Fernandes, + M. G., & Zeviani, W. M. (2012). Impacto de diferentes níveis de + desfolha artificial nos estádios fenológicos do + algodoeiro. Revista de Ciências Agrárias, 35(1), 163–172. + +Zeviani, W. M., Ribeiro, P. J., Bonat, W. H., Shimakura, S. E., & + Muniz, J. A. (2014). The Gamma-count distribution in the analysis + of experimental underdispersed data. Journal of Applied + Statistics, 41(12), + 1–11. http://doi.org/10.1080/02664763.2014.922168 +} +\keyword{datasets} + -- GitLab