qp_pi {predint} | R Documentation |
Simple uncalibrated prediction intervals for quasi-Poisson data
Description
qp_pi()
is a helper function that is internally called by quasi_pois_pi()
. It
calculates simple uncalibrated prediction intervals for Poisson
data with constant overdispersion (quasi-Poisson assumption).
Usage
qp_pi(
newoffset,
histoffset,
lambda,
phi,
q = qnorm(1 - 0.05/2),
alternative = "both",
newdat = NULL,
histdat = NULL,
algorithm = NULL
)
Arguments
newoffset |
number of experimental units in the future clusters |
histoffset |
number of experimental units in the historical clusters |
lambda |
overall Poisson mean |
phi |
dispersion parameter |
q |
quantile used for interval calculation |
alternative |
either "both", "upper" or "lower"
|
newdat |
additional argument to specify the current data set |
histdat |
additional argument to specify the historical data set |
algorithm |
used to define the algorithm for calibration if called via
|
Details
This function returns a simple uncalibrated prediction interval
with as the number of experimental units in the
future clusters,
as the estimate for the Poisson mean obtained from the
historical data,
as the estimate for the dispersion parameter
and
as the number of experimental units per historical cluster.
The direct application of this uncalibrated prediction interval to real life data
is not recommended. Please use the quasi_pois_pi_pi()
functions for real life applications.
Value
qp_pi
returns an object of class c("predint", "quasiPoissonPI")
.
Examples
# Prediction interval
qp_pred <- qp_pi(newoffset=3, lambda=3, phi=3, histoffset=1:9, q=qnorm(1-0.05/2))
summary(qp_pred)