Greetings,
Version 0.1.4-0 of the ChainLadder package is available via CRAN and
http://code.google.com/p/chainladder/
The ChainLadder package, which is focused on claims reserving methods
typically carried out by property/casualty insurance actuaries, has recently
been enhanced to implement the methods in David Clark's 2003 CAS (Casualty
Actuarial Society) *Forum* paper "LDF Curve-Fitting and Stochastic
Reserving: A Maximum Likelihood Approach." Clark's methods are ready to
be
put to use on a wide variety of "triangles".
To see a demonstration of the example in Clark's paper (ChainLadder's
"GenIns" is the same dataset), run
> library(ChainLadder)
> demo(clarkDemo)
This will run the "LDF Method" with the loglogistic growth function
limited
to 20 years of development, per Clark's example. It will also run the
"CapeCod Method" with the weibull growth function. In both cases, per
Clark's recommendation, the option is selected to fit the growth function to
the average date of loss of each accident year (a.k.a., year of origin). The
demo output consists of two displays: the table of expected values, standard
errors and CV's as on page 65 of the paper, and four residual plots.
Type ?ClarkLDF and ?ClarkCapeCod for help on running Clark's two methods.
Please contact the author, Daniel Murphy at chiefmurph _at_ yahoo.com, with
questions, suggestions, ideas or problems.
If anyone is interested in collaborating to develop additional ChainLadder
package functionality, do not hesitate to contact Dan or any of the other
ChainLadder authors. You don't have to be an R expert or an actuary to be
helpful!
More details:
Version 0.1.4-0
===============NEW FEATURES
o New implementation of the methods in David Clark's "LDF Curve
Fitting"
paper in the 2003 Forum by Daniel Murphy.
- Includes LDF and CapeCod methods (functions 'ClarkLDF' and
'ClarkCapeCod', respectively)
- Programmed to handle loglogistic and weibull growth functions
- Printing an object returned by the function results in a table
similar to that on p. 65 of the paper
- Plotting such an object results in four residual plots, including
a Q-Q plot with the results of the Shapiro-Wilk test
Cheers,
The ChainLadder Team
[[alternative HTML version deleted]]
