sparge.modsel {mmodely} | R Documentation |
Coeficients distribution [sparge] plot of models selected from each subset
Description
Plot the raw distribution of points corresponding to the coefficients harvested from the best model of each subset of the dataset.
Usage
sparge.modsel(PC, jit.f=1, R2x=3, nx=2, n.max=max(unlist(PC$n)), zeroline=TRUE,
add=FALSE, pd=0, pvs=names(PC$coefs), pvlabs=NULL,
xlim=range(unlist(PC$coefs)),
MA = NULL, ap=8, ac = 1, ax = nx, ...)
Arguments
PC |
a list of vectors of pooled coefficients (or scores) harvested from the 'best' selected modeling runs (out put from 'get.pgls.coefs') |
jit.f |
factor for random jittering (see 'jitter()' |
R2x |
the line width expansion factor according to R^2 value |
nx |
the point size expansion factor according to sample size of model |
n.max |
the maximum sample size used in all models |
zeroline |
should we add an abline at x=0? |
add |
should we add to the existing plot? |
pd |
'position dodge' moves all y axis plotting positions up or down by this provided value (useful for adding multiple distributions for the same param) |
pvs |
the predictor variable vector for ordering the y-axis labels |
pvlabs |
the predictor variable labels for labeling the plot (defaults to pvs) |
xlim |
x axis plot limits |
MA |
matrix of model averages (defaults to NULL) |
ap |
coded numeric point character symbol used for model averaged parameter position |
ac |
color symbol used for model averaged parameters plot character |
ax |
expansion factor to expant model average parameter plot character (defaults to nx) |
... |
other parameters passed on to plot |
Value
a 'sparge' [sprinkle/smear] plot of coefficent distributions
See Also
See also 'boxplot' and 'stripchart' in package 'graphics' as well as 'violin', 'bean', 'ridgelines', and 'raincloud' plots.
Examples
data.path <- system.file("extdata","primate-example.data.csv", package="mmodely")
data <- read.csv(data.path, row.names=1)
pvs <- names(data[3:5])
data$gn_sp <- rownames(data)
tree.path <- system.file("extdata","primate-springer.2012.tre", package="mmodely")
phyl <- ape::read.tree(tree.path)[[5]]
mods <- get.model.combos(predictor.vars=pvs, outcome.var='OC', min.q=2)
PGLSi <- pgls.iter(models=mods, phylo=phyl, df=data, k=1,l=1,d=1)
coefs.objs <- get.pgls.coefs(PGLSi$fits, est='Estimate')
sparge.modsel(coefs.objs)