famuvie
commented
9 years ago Since the users can extract the raw values to build the variograms, I'd rather let further visualizations to them. Here is an example:
## Example: fit two models to the globulus dataset
data(globulus)
gen.globulus <- list(model = 'add_animal',
pedigree = globulus[,1:3],
id = 'self')
res.blk <- remlf90(fixed = phe_X ~ gg,
genetic = gen.globulus,
spatial = list(model = 'blocks',
coord = globulus[, c('x','y')],
id = 'bl'),
data = globulus)
res.ar <- remlf90(fixed = phe_X ~ gg,
genetic = gen.globulus,
spatial = list(model = 'AR',
coord = globulus[, c('x','y')],
rho = c(.85, .8)),
data = globulus)
## Get the corresponding variograms of residuals
vb <- variogram(res.blk)
va <- variogram(res.ar)
## Extract both isotropic variograms to be compared
iv <- rbind(cbind(model = 'blocks', vb[['isotropic']]),
cbind(model = 'AR1xAR1',va[['isotropic']]))
## Then, you can plot them together under the same scale
ggplot(iv, aes(distance, variogram)) +
geom_point() +
geom_line() +
stat_smooth(se = FALSE, method = 'auto') +
facet_wrap(~ model)
VS requests to have some degree of control over the scale of the variogram. Useful, for example, to compare the variograms from two different models under the same scale.