sqlhexbin: Scatterplots for large surveys

In sqlsurvey: analysis of very large complex survey samples (experimental)

Description

Ordinary scatterplots are not useful for large, unequally sampled data. Two alternatives are hexagonal binning, in which points are replaced by hexagons indicating the population weight in regions of the plot, and thinning, in which one or more iid subsamples taken from the survey sample is plotted.

Usage

## S3 method for class 'sqlsurvey'
svyplot(formula, design, style = c("hex", "grayhex", "subsample"), ...)
## S3 method for class 'sqlrepsurvey'
svyplot(formula, design, style = c("hex", "grayhex", "subsample"), ...)
sqlhexbin(formula, design, xlab = NULL, ylab = NULL, ..., chunksize =
10000)
sqlscatter(formula, design, npoints = 1000, nplots = 4, ...) 

Arguments

formula	Formula specifying the two variables to plot
design	sqlsurvey object
xlab	x-axis label
ylab	y-axis label
style	"hex" does hexagonal binning, "subsample" does thinned scatterplots
npoints	Number of points to sample
nplots	Number of thinned samples to plot
...	Other arguments to gplot.hexbin or plot
chunksize	Number of observations to fetch at a time

Details

Unlike most functions for sqlsurvey objects, sqlhexbin fetches all observations for the specified variables, and so would be slow over a network link.

Value

A object of S4 class hexbin, invisibly.

References

Carr, D. B. et al. (1987) Scatterplot Matrix Techniques for Large N. JASA 83, 398, 424-436.

See Also

svysmooth.sqlsurvey

sqlsurvey documentation built on May 2, 2019, 4:53 p.m.