scattersmooth: Two dimensional Smooth scatter plots

In mstasinopoulos/GAMLSS-utilities: Generalised Additive Models for Location Scale and Shape Utilities

Two dimensional Smooth scatter plots

Description

The function produced two dimensional smooth scatter plots. The method used is described in Eilers and Goeman (2004).

Usage

scattersmooth(x, y, nbin = 100, lambda = 1, ndot = 500,
              csize = 0.3, ticks = TRUE, xlim = c(min(x), 
              max(x)), ylim = c(min(y), max(y)), show = TRUE, 
              save = FALSE, data = NULL, xlab = NULL, 
              ylab = NULL,  cols = heat.colors(10:200), 
              col.points = "blue", ...)

Arguments

x	the x-variable
y	the y-variable
nbin	the number of bins required for smoothing
lambda	the smoothing parameter
ndot	how many data points to show in the plot
csize	the size of the data points
ticks	whether ticks in the x and y axis appear in the plot
xlim	the x limit
ylim	the y limit
show	whether to show the graph or not
save	whether to save the output as a list or not
data	the data file data
xlab	the x label as character string
ylab	the y label as character string
cols	for changing the color scheme, the defaul is heat.colors(10:200). Other suggestions are gray(0:100/100), heat.colors(101), rainbow(100:200), terrain.colors(101), topo.colors(101), cm.colors(101). Note that if you have the package colorspace in R you can used heat_hcl(100) which was the default before.
col.points	the colours of the points
...	for extra arguments

Details

The function is similar to the function smoothScatter() in graphics but it used penelized bin smoother as described in Eilers and Goeman (2004) rather than kernel smoother.

Value

the function produces a two dimensional smooth plot and saves if save=TRUE a list with the following components:

Hraw	A nbin by nbin matrix containing the bin row data
Hsmooth	A nbib by nbib matrix containing the smooth two dimensional histogram
xgrid	the x-grid
ygrid	the y-grid
xbin	the bin for x values
ybin	the bin for y values
nmiss	number of missing values
seldots	the values of the plotted dots

Author(s)

Paul Eilers [e-mail]

References

Eilers, P. H. C. and Goeman, J. J. (2004). Enhancing scatterplots with smoothed density. Bioinformatics, Vol 20 no 5, pp 623-628.

Rigby, R. A. and Stasinopoulos D. M. (2005). Generalized additive models for location, scale and shape,(with discussion), Appl. Statist., 54, part 3, pp 507-554.

Stasinopoulos D. M. Rigby R.A. (2007) Generalized additive models for location scale and shape (GAMLSS) in R. Journal of Statistical Software, Vol. 23, Issue 7, Dec 2007, http://www.jstatsoft.org/v23/i07.

See Also

smoothScatter,gamlss

Examples

m <- 1000
set.seed(pi)
phi <- 2 * pi * runif(m)
rho <- rchisq(m, df = 6)
x <- cos(phi) * rho
y <- sin(phi) * rho
H <- scattersmooth(x, y)
H1 <- scattersmooth(x, y, cols=rainbow(100:200)) 
#  If you have the package colorspace use instead 
# library(colorspace)
# H <- scattersmooth(x, y, cols=heat_hcl(100))
# H1 <- scattersmooth(x, y, cols=rainbow_hcl(100))
data(db)
scattersmooth(age, head,  data=db, cols=terrain.colors(101), ndot=2000, lambda=1)
# or if you have colorspace
#scattersmooth(age, head,  data=db, cols=terrain_hcl(100), ndot=2000, lambda=1)

mstasinopoulos/GAMLSS-utilities documentation built on July 8, 2023, 11:25 p.m.