| predict.Pmix {REBayes} | R Documentation |
Predict Method for Pmix
Description
Predict Method for Poisson Mixtures
Usage
## S3 method for class 'Pmix'
predict(object, newdata, Loss = 2, newexposure = NULL, ...)
Arguments
object |
fitted object of class "Pmix" |
newdata |
Values at which prediction is desired |
Loss |
Loss function used to generate prediction. Currently supported values: 2 to get mean predictions, 1 to get harmonic mean predictions, 0 to get modal predictions or any tau in (0,1) to get tau-th quantile predictions. The posterior harmonic mean is the Bayes rule for quadratic loss weighted by variances as in Clevenson and Zidek (1975). |
newexposure |
exposure values for the predictions |
... |
optional arguments to predict |
Details
The predict method for Pmix objects will compute means, quantiles or
modes of the posterior according to the Loss argument. Typically,
newdata would be passed to predict
Value
A vector of predictions
Author(s)
Jiaying Gu and Roger Koenker
References
Clevenson, M. L. and Zidek, J. V. 1975. Simultaneous Estimation of the Means of Independent Poisson Laws, Journal of the American Statistical Association, 70, 698-705.