# grid_jitter: Remove over-plotting by gridding point data intelligently In geotheory/gridPoints: Functions for gridding point data

## Description

Data jittering reduces overplotting by adding small variances to values. grid_jitter removes it entirely by fitting points to a custom grid. This function applies Hungarian algorithm to match entire points set to the whole grid. (By contrast grid_jitter2 applies Hungarian with small area constraints.) This function works well for smaller grids and point sets, but will slow considerably with larger sets.

## Usage

 ```1 2``` ```grid_jitter(x, y = NULL, nx = 50, ny = NULL, tol = 5, plotresults = TRUE, file = NULL, w = 20, h = 20) ```

## Arguments

 `x` Numeric vector or 2 column matrix or data.frame of data points to plot `y` Numeric vector, y coordinates matching x (if a vector) `nx` Numeric, grid x/y dimensions `ny` Numeric `tol` Numeric, the maximum distance overplotted points are allowed to move to the nearest vacant grid cell `plotresults` Logical, output plots to illustrate point displacements and cell-reallocations `file` String, if not NULL the result of plotresults will save to filename instead of rendering in R `w` Numeric, width/height (inches) of plotresults if output to file (PDF) `h` Numeric

## Value

A 2 column matrix of grid-jittered point coordinates

## Examples

 ``` 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25``` ```# 1. normal distribution d1 = data.frame(x = rnorm(300), y = rnorm(300)) d1g = grid_jitter(d1, nx=50, ny=50, tol=5) # 2. faithful dataset d2 = faithful d2g = grid_jitter(d2, nx=50, ny=50, tol=3) # 3. mpg example d3 = ggplot2::mpg[,c(3,8)] d3g = grid_jitter(d3, nx=100, ny=100, tol=3) # 4. US States d4 = data.frame(x=state.center\$x, y=state.center\$y, id=state.abb) d4j = grid_jitter(d4[,1:2], nx=10, ny=8, tol=3, plotresults=TRUE) cols = sample(colorRampPalette(c('darkblue','blue','lightblue'))(50)) par(mai=rep(.6,4), mfrow=c(1,2)) plot(d4[,1:2], pch=15, cex=2.5, col=cols, asp=2.5, xlab=NA, ylab=NA) text(d4[,1:2], labels=state.abb, col='white', cex=.7) plot(d4j, pch=15, cex=2.5, col=cols, asp=2.5, xlab=NA, ylab=NA) text(d4j, labels=d4\$id, col='white', cex=.7) # 5. to illuste cell reallocation d5 = data.frame(x = c(1,rep(50, 398), 100), y = c(1,rep(50, 398), 100)) d5g = grid_jitter(d5, nx=30, ny=30, tol=15) ```

geotheory/gridPoints documentation built on May 17, 2019, 1:15 a.m.