Resolving build errors in anticipation of submitting to CRAN

Moved gf graphics functions to gtilde package. But fmodel remains here. Added a gmodel as alias for fmodel --- gmodel makes more sense as a name.

Author: dtkaplan

DESCRIPTION

@@ -7,21 +7,18 @@ Maintainer: Daniel Kaplan <kaplan@macalester.edu>
 Description: Functions providing a formula-based interface to ggplot.
 License: MIT + file LICENSE
 LazyData: TRUE
+Depends:
+    R (>= 3.1), ggplot2
 Imports:
-    shiny,
-    miniUI,
-    ggplot2,
-    ggmap,
     lazyeval,
     rpart,
     mosaic,
     magrittr
 RoxygenNote: 5.0.1
 Suggests:
+    dplyr,
     testthat,
     mosaicData,
-    mosaicMapShapes,
-    rpart.plot,
     knitr,
     rmarkdown
 VignetteBuilder: knitr

NAMESPACE

@@ -1,53 +1,30 @@
 # Generated by roxygen2: do not edit by hand
 
-S3method(data_from_model,lm)
-S3method(explanatory_vars,bootstrap_ensemble)
-S3method(explanatory_vars,lm)
 S3method(print,bootstrap_ensemble)
-S3method(response_var,bootstrap_ensemble)
-S3method(response_var,lm)
 export(collinearity)
-export(create_eval_levels)
 export(cv_pred_error)
-export(data_from_model)
 export(effect_size)
 export(ensemble)
 export(evaluate_model)
-export(explanatory_vars)
-export(fborders)
-export(fmap)
 export(fmodel)
-export(formula_gg)
-export(formula_to_gg)
-export(gf_abline)
-export(gf_bar)
-export(gf_boxplot)
-export(gf_counts)
-export(gf_density)
-export(gf_density_2d)
-export(gf_frame)
-export(gf_freqpoly)
-export(gf_hex)
-export(gf_histogram)
-export(gf_hline)
-export(gf_jitter)
-export(gf_line)
-export(gf_path)
-export(gf_point)
-export(gf_text)
-export(gghelper)
-export(kfold_trial)
+export(gmodel)
 export(reference_values)
-export(response_var)
 export(typical_levels)
-import(ggmap)
-import(ggplot2)
-import(miniUI)
-import(shiny)
+importFrom(ggplot2,aes_string)
 importFrom(ggplot2,facet_grid)
 importFrom(ggplot2,facet_wrap)
-importFrom(ggplot2,ggtitle)
-importFrom(ggplot2,xlab)
-importFrom(ggplot2,xlim)
+importFrom(ggplot2,geom_errorbar)
+importFrom(ggplot2,geom_line)
+importFrom(ggplot2,geom_point)
+importFrom(ggplot2,geom_ribbon)
+importFrom(ggplot2,ggplot)
+importFrom(ggplot2,label_both)
 importFrom(ggplot2,ylab)
-importFrom(ggplot2,ylim)
+importFrom(stats,formula)
+importFrom(stats,lm)
+importFrom(stats,median)
+importFrom(stats,model.matrix)
+importFrom(stats,predict)
+importFrom(stats,quantile)
+importFrom(stats,sd)
+importFrom(utils,data)

R/AARP.R

@@ -12,8 +12,8 @@
 #' changes as the insured person ages.) 
 #' 
 #' @docType data
-#'
-#' @usage data(AARP, package = "statisticalModeling")
+#' @name AARP
+#' @usage data(AARP)
 #'
 #' @keywords datasets
 #' 

R/Alder.R

@@ -47,7 +47,7 @@
 
 #'
 #' @docType data
-#'
+#' @name Alder
 #' @usage data(Alder)
 #'
 #' @keywords datasets

R/Birth_weight.R

@@ -6,18 +6,16 @@
 #' weight, and whether the mother smoked is also recorded.  
 #' The data were present by Nolan and Speed to address the question 
 #' of whether there is a link between maternal smoking and the 
-#' baby's health. (See D. Nolan and T.P. Speed (2009) "Stat Labs: Mathematical
-#' Statistics Through Applications")
+#' baby's health.
 #' 
 #' @docType data
-#'
-#' @usage data(Birth_weight, package = "statisticalModeling")
+#' @name Birth_weight
+#' @usage data(Birth_weight)
 #'
 #' @keywords datasets
 #' 
-#' @source The "AARP Bulletin". 
-#' A copy of the ad is available \href{http://tiny.cc/mosaic/AARP-insurance-ad.pdf}{at this link.}
-#'
+#' @source D. Nolan and T.P. Speed (2009) "Stat Labs: Mathematical
+#' Statistics Through Applications"
 #' @format
 #'   A data frame with 886 observations on the following variables.
 #'   \itemize{

R/College_grades.R

@@ -8,8 +8,8 @@
 #' are included. Only courses with 10 or more students enrolled were included.
 #' 
 #' @docType data
-#'
-#' @usage data(College_grades, package = "statisticalModeling")
+#' @name College_grades
+#' @usage data(College_grades)
 #'
 #' @keywords datasets
 #' 
@@ -34,5 +34,7 @@
 #'   }
 #' 
 #' @examples
-#' GPA <- mosaic::mean(gradepoint ~ sid, data = College_grades)
+#' \dontrun{
+#' GPA <- lm(gradepoint ~ sid - 1, data = College_grades)
+#' }
 "College_grades"

R/Crime.R

@@ -6,7 +6,7 @@
 #' 
 #' @docType data
 #' @name Crime
-#' @usage data(Crime, package = "statisticalModeling")
+#' @usage data(Crime)
 #'
 #' @keywords datasets
 #'

R/HDD_Minneapolis.R

@@ -0,0 +1,35 @@
+#' Heating degree days in Minneapolis, Minnesota, USA
+#'
+#' A "heating degree day" is a measure of weather coldness. It's defined to be the difference
+#' between the outdoor ambient temperature and 65 degrees F, but has a value of zero when the ambient 
+#' temperature is above 65 degrees. This difference is averaged over time and multiplied by the number 
+#' of days in the time period covered. The heating degree day is often used as a measure of the demand for 
+#' domestic heating in a locale.
+#' 
+#' These data report monthly heating degree days. For teaching purposes, the data give an extreme example 
+#' of how a relationship (hdd vs year) can be revealed by including a covariate (month). Although interest focusses
+#' on the change in temperature over the century the data cover, there is such regular seasonal variation that 
+#' no systematic trend over the years is evident unless month is taken into account.
+#' 
+#' @format A data frame \code{HDD_Minneapolis} with 1412 rows and 4 variables:
+#' \itemize{
+#' \item year the year
+#' \item month the month
+#' \item hdd the number of heating degree days for that period.
+#' \item loc the location at which the temperature was measured. In the early years, this was downtown Minneapolis.
+#' Later, the site was moved to the Minneapolis/Saint-Paul International Airport.
+#' }
+
+
+#'
+#' @docType data
+#' @name HDD_Minneapolis
+#' @usage data(HDD_Minneapolis)
+#'
+#' @keywords datasets
+#'
+
+#' @examples
+#' mod_1 <- lm(hdd ~ year, data = HDD_Minneapolis)
+#' mod_2 <- lm(hdd ~ year + month, data = HDD_Minneapolis)
+"HDD_Minneapolis"

R/Houses_for_sale.R

@@ -3,14 +3,14 @@
 #' A random sample of 1,728 homes taken from public records 
 #' from the Saratoga County 
 #' (http://www.saratogacountyny.gov/departments/real-property-tax-service-agency/).
-#' Collected by Candice Corvetti (Williams College ’07) for her senior thesis.
+#' Collected by Candice Corvetti (Williams College '07) for her senior thesis.
 #' 
 #' These data are part of a case study developed by Prof. Dick de Veaux at 
 #' Williams. They are available from the American Statistical Association's \href{http://community.amstat.org/stats101/home}{Stat 101} collection
 #' of case studies and included in this package for convenience.
 #'
 #' @docType data
-#'
+#' @name Houses_for_sale
 #' @usage data(Houses_for_sale)
 #'
 #' @format A dataframe with 1728 cases, each of which is a house for sale. 
@@ -19,7 +19,7 @@
 #'
 #' @references Dick De Veaux (2015) "How much is a fireplace worth?" Stats 101: A resource for teaching introductory statistics, American Statistical Association
 #'
-#' @source \href{http://community.amstat.org/stats101/home}{Stat 101}
+# @source \href{http://community.amstat.org/stats101/home}{Stat 101}
 #' 
 #' @examples
 #' mod_1 <- lm(price ~ fireplaces, data = Houses_for_sale)

R/Mussels.R

@@ -0,0 +1,46 @@
+#' Metabolism of zebra mussels
+#'
+#' Zebra mussels are a small, fast reproducing species of freshwater mussel 
+#' native to the lakes of southeast Russia. They have accidentally been introduced in other 
+#' areas, competing with native species and creating problems for people as they cover the 
+#' undersides of docks and boats, clog water intakes and other underwater structures. 
+#' Zebra mussels even attach themselves to other mussels, sometimes starving those mussels.
+#' 
+#' Ecologists Shirley Baker and Daniel Hornbach examined whether zebra mussels gain an advantage by attaching to other mussels rather than to rocks.(baker-hornbach-2008) The ecologists collected samples of small rocks and Amblema plicata mussels, each of which had a collection of zebra mussels attached. The samples were transported to a laboratory where the group of mussels from each individual rock or Amblema were removed and placed in an aquarium equipped to measure oxygen uptake and ammonia excretion. After these physiological measurements were made, the biochemical composition of the mussel tissue was determined: the percentage of protein, lipid, carbohydrate, and ash.
+
+#' Baker and Hornbach found that zebra mussels attached to Amblema had greater 
+#' physiological activity than those attached to rocks as measured by oxygen uptake and 
+#' ammonia excretion. But this appears to be a sign of extra effort for the Amblema-attached 
+#' zebra mussels, since they had lower carbohydrate and lipid levels. In other words, attaching to 
+#' Amblema appears to be disadvantageous to the zebra mussels compared to attaching to a rock.
+
+#' 
+#' @format A data frame \code{Mussels} with 30 rows and 11 variables. 
+#'   \itemize{
+#'     \item{\code{GroupID}} {ID for the cluster of mussels growing on a substrate. }
+#'     \item{\code{dry.mass}} {The mass of the mussels (as a group) after dehydration.}
+#'     \item{\code{count}} {How many mussels were in the cluster.}
+#'     \item{\code{attachment}} {The substrate to which the mussels were attached.}
+#'     \item{\code{lipid}} {Percentage of dry mass that is lipid.}
+#'     \item{\code{protein}} {Percentage of dry mass that is protein.}
+#'     \item{\code{carbo}} {Percentage of dry mass that is carbohydrate.}
+#'     \item{\code{ash}} {Percentage of dry mass that is ash.}
+#'     \item{\code{ammonia}} {Nitrogen excretion measured as ammonia in mg per hour for the group.}
+#'     \item{\code{Kcal}} {Total calorific value of the tissue in kilo-calories per gram.}
+#'     \item{\code{O2}} {Oxygen uptake in mg per hour for the group.}
+#'   }
+#'   
+#' @docType data
+#' @name Mussels
+#' @usage data(Mussels)
+#'
+#' @keywords datasets
+#'
+
+#' @examples
+#' Mussels$ind.mass <- with(Mussels, dry.mass/count)
+#' mod_1 <- lm(O2/count ~ attachment, data = Mussels)
+#' mod_2 <- lm(ammonia/count ~ attachment, data = Mussels)
+#' mod_3 <- lm(O2/count ~ ind.mass + attachment, data = Mussels)
+#' mod_4 <- lm(ammonia/count ~ ind.mass + attachment, data = Mussels)
+"Mussels"