| missingDataGrid {aqp} | R Documentation |
Generate a levelplot of missing data from a SoilProfileCollection object.
missingDataGrid(s, max_depth, vars, filter.column = NULL, filter.regex = NULL, cols = NULL, ...)
s |
a SoilProfilecollection object |
max_depth |
integer specifying the max depth of analysis |
vars |
character vector of column names over which to evaluate missing data |
filter.column |
a character string naming the column to apply the filter REGEX to |
filter.regex |
a character string with a regular expression used to filter horizon data OUT of the analysis |
cols |
a vector of colors |
... |
additional arguments passed on to |
This function evaluates a 'missing data fraction' based on slice-wise evaulation of named variables in a SoilProfileCollection object.
A data.frame describing the percentage of missing data by variable.
A lattice graphic is printed to the active output device.
D.E. Beaudette
## visualizing missing data
# 10 random profiles
require(plyr)
s <- ldply(1:10, random_profile)
# randomly sprinkle some missing data
s[sample(nrow(s), 5), 'p1'] <- NA
s[sample(nrow(s), 5), 'p2'] <- NA
s[sample(nrow(s), 5), 'p3'] <- NA
# set all p4 and p5 attributes of `soil 1' to NA
s[which(s$id == '1'), 'p5'] <- NA
s[which(s$id == '1'), 'p4'] <- NA
# upgrade to SPC
depths(s) <- id ~ top + bottom
# plot missing data via slicing + levelplot
missingDataGrid(s, max_depth=100, vars=c('p1', 'p2', 'p3', 'p4', 'p5'),
main='Missing Data Fraction')
## id p1 p2 p3 p4 p5
## 1 1 33 0 0 100 100
## 2 2 0 33 0 0 0
## 3 3 17 17 0 0 0
## 4 4 0 0 33 0 0
## 5 5 0 0 0 0 0
## 6 6 0 20 40 0 0
## 7 7 0 0 17 0 0
## 8 8 17 17 0 0 0
## 9 9 33 0 33 0 0
## 10 10 17 17 0 0 0