Open Access. Powered by Scholars. Published by Universities.®
Articles 1 - 3 of 3
Full-Text Articles in Insurance
Ultimate Ruin Probability For A Time-Series Risk Model With Dependent Classes Of Insurance Business, Lai Mei Wan, Kam Chuen Yuen, Wai Keung Li
Ultimate Ruin Probability For A Time-Series Risk Model With Dependent Classes Of Insurance Business, Lai Mei Wan, Kam Chuen Yuen, Wai Keung Li
Journal of Actuarial Practice (1993-2006)
We consider a discrete-time risk model with m (m ~ 2) dependent classes of insurance business. The claim processes of these m classes are assumed to follow a multivariate autoregressive time-series model of order 1. Given this claims model, we explore the probability of ultimate ruin assuming exponentially bounded claims. As an example, we use simulations to study the case where there are two business and the underlying losses are of two types: bivariate exponential and bivariate gamma claim distributions.
Life Contingencies With Stochastic Discounting Using Moving Average Models, Steven Haberman, Russell Gerrard, Dimitrios Velmachos
Life Contingencies With Stochastic Discounting Using Moving Average Models, Steven Haberman, Russell Gerrard, Dimitrios Velmachos
Journal of Actuarial Practice (1993-2006)
This paper offers simplified procedures for calculating moments of functions in life contingencies when the random force of interest is modeled using an unconditional moving average process of order q, MA(q). It extends the MA(l) model that has been used for stochastic discounting. Using the more general MA(q) model allows actuaries to better capture the auto correlation between successive interest rates in a time series.
Constrained Forecasting Of The Number Of Ibnr Claims, Louis G. Doray
Constrained Forecasting Of The Number Of Ibnr Claims, Louis G. Doray
Journal of Actuarial Practice (1993-2006)
We consider the problem of forecasting the number of claims incurred. After subtracting the number of claims reported to date, the number of claims incurred but not reported (IBNR) can be forecasted. The basic model assumes that the number of claims per accident period follows an autoregressive moving average time series process. Instead of assuming the data are available in the usual claim run-off triangle format, we assume that the only data available are the number of claims reported at the valuation date for each accident interval of an observation period. Box-Jenkins methods are used to forecast the ultimate number …