The Lee-Carter model under Bayesian analysis : Markov Chain Monte Carlo simulation with WinBUGS.

Mortality improvement has become a significant concern in mortality projections because it directly affects the risk in life insurance business. As such, the Lee-Carter model that uses a stochastic framework is preferred against other deterministic models because of its allowance for the associated...

Full description

Saved in:
Bibliographic Details
Main Authors: Tan, Chong It., Lu, Xin., Chong, Zi Kent.
Other Authors: Li Ka Ki Jackie
Format: Final Year Project
Language:English
Published: 2010
Subjects:
Online Access:http://hdl.handle.net/10356/33720
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:Mortality improvement has become a significant concern in mortality projections because it directly affects the risk in life insurance business. As such, the Lee-Carter model that uses a stochastic framework is preferred against other deterministic models because of its allowance for the associated uncertainty. Various estimation methods have been proposed to estimate its parameters. In this paper, we consider the Lee-Carter model under the context of Bayesian analysis, which is able to furnish a posterior distribution for each parameter or variable of interest. We propose using a method called Markov Chain Monte Carlo (MCMC) simulation to estimate the parameters in the Lee-Carter model. Specifically, we use WinBUGS software to carry out the parameter estimation.