Merge pull request #26 from USEPA/develop

Develop

.Rbuildignore

@@ -6,4 +6,4 @@
 ^Meta$
 ^\.github$
 ^LICENSE\.md$
-^NEWS.md$
+^notes.md$

.gitignore

@@ -5,3 +5,4 @@ spsurvey.Rproj
 inst/doc
 /doc/
 /Meta/
+notes.md

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: spsurvey
 Title: Spatial Sampling Design and Analysis
-Version: 5.0.1
+Version: 5.1.0
 Authors@R: c(
     person("Michael", "Dumelle", role=c("aut","cre"),
            email = "[email protected]", comment = c(ORCID = "0000-0002-3393-5529")),
@@ -27,6 +27,7 @@ Imports:
     grDevices,
     lme4,
     MASS,
+    sampling,
     stats,
     units
 Suggests: 
@@ -39,5 +40,5 @@ BugReports: https://github.com/USEPA/spsurvey/issues
 VignetteBuilder: knitr
 Encoding: UTF-8
 LazyData: true
-RoxygenNote: 7.1.1
+RoxygenNote: 7.1.2
 NeedsCompilation: no

NAMESPACE

@@ -18,6 +18,9 @@ export(diffrisk_analysis)
 export(errorprnt)
 export(grts)
 export(irs)
+export(localmean_cov)
+export(localmean_var)
+export(localmean_weight)
 export(pd_summary)
 export(power_dsgn)
 export(ppd_plot)
@@ -54,6 +57,7 @@ importFrom(graphics,text)
 importFrom(graphics,title)
 importFrom(lme4,VarCorr)
 importFrom(lme4,lmer)
+importFrom(sampling,UPpivotal)
 importFrom(stats,addmargins)
 importFrom(stats,chisq.test)
 importFrom(stats,confint)

NEWS.md

@@ -1,3 +1,22 @@
+# spsurvey 5.1.0 (2021-12-14)
+
+## Minor updates
+
+* Added a `Total` option to the `statistics` argument in `cont_analysis()`.
+* Added `localmean_weight()`, `localmean_var()`, and `localmean_cov()` functions to compute the local neighborhood variance estimator outside of the `*_analysis()` functions.
+* Added an option to provide a bounding box vector to the `fix_bbox` argument in `sp_plot()`.
+* Added an error message to `grts() ` and `irs()` that stops the function when geographic coordinates are used.
+* Added an error message to `grts()` and `irs()` that stops the function when too many expected samples are specified for at least one level of an unequal probability variable.
+* Added a `caty_n_over` argument to `grts()` and `irs()` that makes it easier to specify reverse hierarchically ordered replacement sites for unequal probability sampling designs.
+* Added vector argument support for the `n_over` and `n_near` arguments in `grts()` and `irs()` (list arguments were already supported).
+* Added a default value for `siteID` in `attrisk_analysis()`, `cat_analysis()`, `cont_analysis()`, `diffrisk_analysis()`, and `relrisk_analysis()` that assumes each row in `dframe` represents a unique site.
+
+## Bug fixes
+
+* Fixed a bug in `irs()` that made the algorithm sensitive to the ordering of `sframe`.
+* Fixed a bug in percentile estimation from `cont_analysis()` that incorrectly copied estimate values.
+* Fixed a bug in `grts()` and `irs()` that prevented sampling from `sframe` when the geometry type was `LINESTRING`, `MULTILINESTRING`, `POLYGON`, or `MULTIPOLYGON` and the number of desired samples exceeded the number of rows.
+
 # spsurvey 5.0.1 (2021-10-18)
 
 ## Bug fix
@@ -50,7 +69,7 @@ this is reported on the CRAN results pages for fedora-clang and fedora-gcc.
 
 ## Bug fix
 
-* Update test data and Description file to handle changes in `sf` and changes to `crs` described [here](https://www.r-spatial.org/r/2020/03/17/wkt.html)
+* Update test data and Description file to handle changes in `sf` and changes to `crs` described [here](https://r-spatial.org/r/2020/03/17/wkt.html)
 
 # spsurvey 4.1.1 (2019-12-12)
 
@@ -66,7 +85,7 @@ this is reported on the CRAN results pages for fedora-clang and fedora-gcc.
 * The package no longer contains any functions written in C.  The C functions were either no longer required due to the transition to use of `sf` objects to contain the survey frame or were replaced with functions written in R.  The new R functions are: `cellWeight`, `constructAddr`, `insideAreaGridCell`, `insideLinearGridCell`, `make_grid`, `numLevels`, `pickFiniteSamplePoints`, `pickGridCells`, `pickSamplePoints`, `ranho`, and `selectFeatureID`.
 
 ## Bug fixes
-* The new R code function named `numLevels` that determines the number of hierarchical levels for a generalized random-tesselation stratified (GRTS) survey design now includes code to ensure that the maximum number of levels (which is currently 11) is not bypassed when creating a survey design.
+* The new R code function named `numLevels` that determines the number of hierarchical levels for a generalized random-tessellation stratified (GRTS) survey design now includes code to ensure that the maximum number of levels (which is currently 11) is not bypassed when creating a survey design.
 
 ## Modified function
 * Modified function `input.check` to check for missing values among the x-coordinates and y-coordinates for location.

R/attrisk_analysis.R

@@ -54,7 +54,7 @@
 #'
 #' @param response_levels List providing the category values (levels) for each
 #'   element in the \code{vars_response} argument.  Each element in the list
-#'   must contian two values, where the first value identifies poor condition,
+#'   must contain two values, where the first value identifies poor condition,
 #'   and the second value identifies good condition.  This argument must be
 #'   named and must be the same length as argument \code{vars_response}.  Names
 #'   for this argument must match the values in the \code{vars_response}
@@ -85,7 +85,8 @@
 #'
 #' @param siteID Character value providing the name of the site ID variable in
 #'   \code{dframe}.  For a two-stage sample, the site ID variable
-#'   identifies stage two site IDs.  The default value is \code{"siteID"}.
+#'   identifies stage two site IDs.  The default value is \code{NULL}, which
+#'   assumes that each row in \code{dframe} represents a unique site.
 #'
 #' @param weight Character value providing the name of the design weight
 #'   variable in \code{dframe}.  For a two-stage sample, the
@@ -340,7 +341,7 @@
 ################################################################################
 
 attrisk_analysis <- function(dframe, vars_response, vars_stressor, response_levels = NULL,
-                             stressor_levels = NULL, subpops = NULL, siteID = "siteID", weight = "weight",
+                             stressor_levels = NULL, subpops = NULL, siteID = NULL, weight = "weight",
                              xcoord = NULL, ycoord = NULL, stratumID = NULL, clusterID = NULL,
                              weight1 = NULL, xcoord1 = NULL, ycoord1 = NULL, sizeweight = FALSE,
                              sweight = NULL, sweight1 = NULL, fpc = NULL, popsize = NULL,
@@ -394,6 +395,13 @@ attrisk_analysis <- function(dframe, vars_response, vars_stressor, response_leve
 
   dframe <- droplevels(dframe)
 
+  # If no siteID is provided, set one that assumes each row is a unique site
+
+  if (is.null(siteID)) {
+    siteID <- "siteID"
+    dframe$siteID <- paste("site", seq_len(nrow(dframe)), sep = "-")
+  }
+
   # Ensure that the dframe data frame contains the site ID variable
 
   if (!(siteID %in% names(dframe))) {

R/cat_analysis.R

@@ -47,7 +47,8 @@
 #'
 #' @param siteID Character value providing name of the site ID variable in
 #'   the \code{dframe} data frame.  For a two-stage sample, the site ID variable
-#'   identifies stage two site IDs.  The default value is \code{"siteID"}.
+#'   identifies stage two site IDs.  The default value is \code{NULL}, which
+#'   assumes that each row in \code{dframe} represents a unique site.
 #'
 #' @param weight Character value providing name of the design weight
 #'   variable in \code{dframe}.  For a two-stage sample, the
@@ -284,7 +285,7 @@
 #' @export
 ################################################################################
 
-cat_analysis <- function(dframe, vars, subpops = NULL, siteID = "siteID", weight = "weight",
+cat_analysis <- function(dframe, vars, subpops = NULL, siteID = NULL, weight = "weight",
                          xcoord = NULL, ycoord = NULL, stratumID = NULL, clusterID = NULL,
                          weight1 = NULL, xcoord1 = NULL, ycoord1 = NULL, sizeweight = FALSE,
                          sweight = NULL, sweight1 = NULL, fpc = NULL, popsize = NULL,
@@ -338,6 +339,13 @@ cat_analysis <- function(dframe, vars, subpops = NULL, siteID = "siteID", weight
 
   dframe <- droplevels(dframe)
 
+  # If no siteID is provided, set one that assumes each row is a unique site
+
+  if (is.null(siteID)) {
+    siteID <- "siteID"
+    dframe$siteID <- paste("site", seq_len(nrow(dframe)), sep = "-")
+  }
+
   # Ensure that the dframe data frame contains the site ID variable
 
   if (!(siteID %in% names(dframe))) {

R/cdf_plot.R

@@ -101,6 +101,7 @@
 #' @keywords survey
 #'
 #' @examples
+#' \dontrun{
 #' dframe <- data.frame(
 #'   siteID = paste0("Site", 1:100),
 #'   wgt = runif(100, 10, 100),
@@ -135,6 +136,7 @@
 #'   ylab = "Percent of Stream Length", ylab_r = "Same",
 #'   main = "Estimates for Resource Class: Good"
 #' )
+#' }
 #' @export
 ################################################################################
 

R/change_analysis.R

@@ -154,7 +154,7 @@
 #'   named vector containing one more than the number of clusters in the sample,
 #'   where the first item in the vector specifies the number of clusters in the
 #'   population and each subsequent item specifies the number of stage two units
-#'   for the cluster.  The name for the first item in the vector is arbitrry.
+#'   for the cluster.  The name for the first item in the vector is arbitrary.
 #'   Subsequent names in the vector identify clusters and must match the cluster
 #'   IDs.  For a stratified design, the object is a named list of vectors, where
 #'   names must match the strata IDs.  For each stratum, the format of the