Description Usage Arguments Details Value Author(s) See Also Examples
Bayesian Model To Identify Factors Affecting Wildlife-Vehicle Collisions
1 | finalModel(X, vectorFinalVariables, collisions, nYear, Area, departement)
|
X |
a data.frame containing the numeric variables supposed to have an effect on the wildlife-vehicle collisions (columns) for spatial unit (rows). |
vectorFinalVariables |
a vector of character strings containing the names of the final variables used in the combination. |
collisions |
an integer vector with length equal to
|
nYear |
an integer vector with length equal to |
Area |
a numeric vector with length equal to |
departement |
a character vector with length equal to
|
finalModel
fits the final Bayesian model used to
predict the number of collisions between ungulates and vehicle as a
function of a linear combination of a set of environmental
variables.
a list with all elements required for the fit of the model
with JAGS, that is: (i) data4jags
: the list of the data
required by the model, to be passed to the argument data
of
the function jags.model
of the package rjags
, (ii)
ini
: list of starting values for the parameters, to be
passed to the argument init
of the function
jags.model
, (iii) modelstring
: a character string
containing the model fit by JAGS, (iv) coefnames
: vector of
character strings containing the names of the coefficients of
interest in the model, to be passed to the argument
variable.names
of the function coda.samples
of the
package rjags
.
Clement Calenge, clement.calenge@oncfs.gouv.fr
jags.model
,
coda.samples
, prepareFit
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 | ## The data used for the fit
data("dataCollision")
## Consider the Kuo-Mallik model fit (see ?prepareFit to see how
## this model was fitted)
data("modelRedDeer")
## we consider the probability associated to each model
probabilityKM(modelRedDeer)
## The best model seems to be the model including forFrag and hunt
## Prepare the data
X <- dataCollision$RedDeer$X
## We scale the variables to improve mixing
X <- as.data.frame(scale(X))
## Prepare the final model, including only the variables forFrag and hunt
pfm <- finalModel(X, c("forFrag","hunt"), dataCollision$RedDeer$coll,
dataCollision$RedDeer$Y, dataCollision$RedDeer$Area,
dataCollision$RedDeer$departement)
## Not run:
## WARNING: long execution (about 10 min)!!
## the results are stored in the dataset "finalModelRedDeer"
## But if you want to try it, the data are now ready for the fit:
mo <- jags.model(textConnection(pfm$modelstring), ini=pfm$ini, data=pfm$data4jags)
update(mo, n.iter=1000)
finalModelRedDeer <- coda.samples(mo, variable.names = pfm$coefnames,
n.iter = 500000, thin=100)
## End(Not run)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.