-
Notifications
You must be signed in to change notification settings - Fork 9
Open
Description
Message to the Distance list regarding variance estimation when using strip transects with DSM. User first tried dsm_var_prop, but that failed, so tried dsm_var_gam which also failed.
Minimum reproducible example with Spotted dolphin data shipped with dsm package
library(Distance)
library(dsm)
# load the Gulf of Mexico dolphin data (see ?mexdolphins)
data(mexdolphins)
uni.mod <- ds(distdata, key="unif", adjustment = NULL, truncation=6000)
# fit a simple smooth of x and y
dsm.uni <- dsm(count~s(x, y), uni.mod, segdata, obsdata)
# second attempt using dummy_ddf
dsm.uni.dummy <- dsm(count~s(x, y), dummy_ddf(obsdata$object,width=6000), segdata, obsdata)
# Calculate the variance
# this will give a summary over the whole area in mexdolphins$preddata
mod.uni.var <- dsm_var_gam(dsm.uni, preddata, off.set=preddata$area)
summary(mod.uni.var)
mod.uni.var.dummy <- dsm_var_gam(dsm.uni.dummy, preddata, off.set=preddata$area)
summary(mod.uni.var.dummy)
Preferred behaviour
dsm_var_gam should recognise when a strip transect has been used and return simply the variability of the fitted gam because there is no uncertainty associated with the detection function model
Metadata
Metadata
Assignees
Labels
No labels