Nothing
#' @title Point layer of Stores selling Food in Dallas County, TX
#' @description Location of food stores in Dallas County, TX, in the longitude and
#' latitude format (see \code{proj4string=CRS("+proj=longlat +ellps=WGS84")})..
#' @docType data
#' @name foodStoresShp
#' @author The data were compiled by Laila Al Aswad and Michael Tiefelsdorf <tiefelsdorf@@utdallas.edu>
#' @source Reference USA, 2019 \url{http://www.referenceusa.com}.
#' @format Spatial polygon data-frame with 1623 verified store locations.
#' \describe{
#' \item{SALESVOL}{Reported total annual sales volume of goods in $}
#' \item{PROPFOOD}{Assumed proportion of food sales}
#' \item{FOODSALES}{Calculated annual sales volume of food in $}
#' \item{STORETYPE}{Factor distinguishing between stores selling nutritious
#' food (grocery stores) and processed food (convenience
#' stores)}
#' }
#' @examples
#' library(spatstat)
#' library(rgdal)
#' library(sp)
#' proj4string(bndShp) # Current system
#' projUTM <- CRS("+proj=utm +zone=14 +units=m") # isotropic coordinate sytem
#' bndUTM <- spTransform(bndShp, projUTM) # Re-project boundary
#' storesUTM <- spTransform(foodStoresShp, projUTM) # Re-project points
#' storesDf <- as.data.frame(storesUTM) # Extract data-frame
#' storesPts <- as.ppp(storesUTM) # Convert to .ppp
#' storesPts$marks <- NULL # Clear marks
#' bndWin <- as.mask(as.owin(bndUTM), eps=200) # pixel window with 200 m resolution
#' unitname(bndWin) <- list("meter","meters") # set units
#' storesPts <- storesPts[bndWin] # assign window to pts
#' summary(storesPts)
#'
#' ## Evaluate weighted kernel density with bw=3000
#' allFoodIm <- density(storesPts, weights=storesDf$FOODSALES, sigma=3000)
#' plot(allFoodIm, main="All Stores Weighted Kernel Density\nbw = 3000 m")
#' plot(storesPts, cex=0.5, pch=16, col="green", add=TRUE)
#' box(); axis(1); axis(2)
#'
NULL
Any scripts or data that you put into this service are public.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.