show.pred.weed {WeedMap} | R Documentation |
Plot traces of Markov chains simulated by
pred.weed
, compute statistics on the posterior distribution,
compare to known true values. (if any)
show.pred.weed(sim = NULL, res, param = TRUE, pairs = TRUE, wy = TRUE, wz = TRUE, nit, thin, burnin)
sim |
A list, typically the output of function sim.weed |
res |
A list, typically the output of function pred.weed |
param |
Logical: if TRUE graphs of simulated parameters contained in
res are displayed |
pairs |
Logical: if TRUE pair plots of predicted weed
values are ploted against known true values provided as outputof
function sim are displayed (for debuging and convergence checking) |
wy |
Logical: if TRUE chains corresponding to the nine
first sites in y are displayed |
wz |
Logical: if TRUE chains corresponding to the nine
first sites in z are displayed |
nit |
Number of Markov iterations |
thin |
Thinning of the chain |
burnin |
Number of values to discard before computing statistics
about the empirical posterior. The number of iterations discarded is
the number of stored iterations, e.g. if nit=10000 and
thin=10 , then only 1000 iterations are stored and
burnin=100 will remove the 100 first which correspond to
actually thin*burnin=1000 Markov iterations. |
No value returned
Gilles Guillot
G. Guillot, N. Loren, M. Rudemo, Bayesian spatial prediction of weed intensities from exact count data and picture based indexes, 2006, submitted