Actuarial Modeling with MCMC and BUGS

Many of the files in this directory represent the output from a project that was originally supported and peer reviewed by the Actuarial Education and Research Fund. The goal of this project was to examine and illustrate the use of MCMC and BUGS in a variety of actuarial modeling (modelling) contexts. Much of the recent work has been supported by the Natural Sciences and Engineering Research Council of Canada (NSERC). Please contact the researcher attached to this project, David P.M. Scollnik, if you have any questions.

Please Note ...

The *.odc files found on the pages linked to below need to be saved to disk and not viewed. Previously, some browsers have experienced difficulty saving these files to disk as binary type. We believe that this problem has since been repaired from the server side of things. Should you still experience any trap errors when opening one of these *.odc files in WinBUGS, please notify me of this fact.

Actuarial Modeling with MCMC and BUGS.

Download the paper in Pdf. This paper appeared in the April, 2001, issue of North American Actuarial Journal. [Scollnik, D.P.M. (2001). Actuarial Modeling with MCMC and BUGS. North American Actuarial Journal, 5 (2), pages 96-124.] [Minor revisions were last made to the paper on January 19, 2000.]

Claims : A Hierarchical Poisson Model for Claim Frequency Data.

Counts : Modeling Arbitrary Discrete Distributions.

Grouped : Modeling Grouped Size of Loss Data.

Actuarial Modeling with MCMC and BUGS : Additional Worked Examples.

Download the Report in Postscript format or Pdf. (If that Pdf file does not work with your reader, then please try this Pdf file instead.) This report was published in ARCH. [Scollnik, D.P.M. (2000). Actuarial Modeling with MCMC and BUGS: Additional Worked Examples. Actuarial Research Clearing House, 2000.2, pages 433-585.]

Exact : Size of Loss Distributions for Exact Data.

Bivreg : Regression Models for Bivariate Loss Data.

Motor : A Predictive Aggregate Claims Distribution.

Norweg : Heterogeneity in Group Life Insurance.

Credit : An Unbalanced 2-Way Crossed Classification Model.

Health : A Normal Model for Order Restricted Graduation.

Broff : Increasing Ordered Graduation of Mortality Rates.

Kimjo : Kimeldorf-Jones Styled Bayesian Graduations.

Monty : The Monty Hall, or Let's Make a Deal, Problem.

Additional Notes and Papers of Possible Interest.

Poster and Related Papers from the Seventh Valencia International Meeting on Bayesian Statistics.

BUGS Software and Related links :

The original BUGS and WinBUGS website is here.

The new OpenBugs site may be found here.

The WinBUGS Development web-site is here.

Additional BUGS Web Resources are located here.

visits to this page have been made since July 1, 1999.

visits to this page have been made since October 18, 2000.

Some of the files on this site are only available in PostScript format. If your computer is not equipped to view PostScript files, you should obtain a PostScript interpreter and viewer from www.cs.wisc.edu/~ghost/index.html or www.cs.wisc.edu/~ghost/.