GammaTrial {bootstrap} | R Documentation |
This is the second stage of bootstrap sampling.
This function do gamma simulation on each cell (not 0) of the lower part of a trianle. Gamma Distribution on each cell is assumed with: shape=Mij/scale, scale, This is all from the assumption that: each cell value is the mean and scale is got from Pearson scale factor: scale=sum(chi-square) / (r+p). See VE paper for detail.
Gamma distribution has: Mean=Shape * Scale, and Variance=Shape * Scale ^2
GammaTrial(A, scale_, ...)
A |
~~incremental lower developed triangle~~ |
scale_ |
~~bootstrap scale factor~~ |
... |
~~After this gamma sampling, bootstrap process is ready to calculate reserves by adding up the incremental cell values on each year.~~
"Analytic and bootstrap estimates of prediction errors in claims reserving" from Peter England and Richard Verrall
##---- Should be DIRECTLY executable !! ---- ##-- ==> Define data, use random, ##-- or do help(data=index) for the standard data sets. ## The function is currently defined as function(A, scale_, ...) standardGeneric("GammaTrial")