ddhazard
Predict function for the result of ddhazard
# S3 method for fahrmeier_94
predict(object, new_data, type = c("response", "term"),
tstart = "start", tstop = "stop", use_parallel = F, sds = F,
max_threads = getOption("ddhazard_max_threads"), ...)
Result of a ddhazard
call
New data to base predictions on
Either "response"
for predicted probability of death or "term"
for predicted terms in the linear predictor
same as tstart
for the stop argument
TRUE
if computation for type = "response"
should be computed in parallel with the parallel
package
Maximum number of threads to use. -1 if it should be determined by a call to detectCores
Not used
The result of type = "term"
is a list with the following elements
terms
Is a 3D array. The first dimension is the number of bins, the second dimension is rows in new_data
and the last dimension is the state space terms
sds
Similar to terms
for the point wise confidence intervals using the smoothed co-variance matrices
fixed_terms
Vector of the fixed effect terms for each observation
The result of type = "response"
is a list with the elements below. The function check if there are columns in new_data
which's names match tstart
and tstop
. If not, then each row in new data will get a predicted probability of dying in every bin.
fits
Fitted probability of dying
istart
Vector with the index of the first bin the elements in fits
is in
istop
Vector with the index of the last bin the elements in fits
is in