Sensitivity analysis of insurance risk models via simulation

Article Abstract:

Performance indicators associated with insurance risk models are subjected to sensitivity analysis and stochastic optimization tests. The first involves evaluation of derivatives of performance measures with respect to parameters of interest while the second takes into account the decision making process of the entire system. Ruin probabilities, their derivatives and tail probabilities of accumulated claims during a given period are examined given a varying number of claims and claim size. The score function method of sensitivity analysis is used to evaluate distributional parameters of the risk model while the push-out method combined with the score function is applied to structural parameters. The same model is used to develop a reliable estimator of the optimal solution in a problem involving excess-of-loss reinsurance.

Insurance, Insurance Carriers and Related Activities, INSURANCE CARRIERS, Analysis, Economic aspects, Insurance industry, Risk (Economics)

---

Antithetic variates, multivariate dependence and simulation of stochastic systems

Article Abstract:

Variance reductions of simulation in stochastic systems with dependent variables are discussed from both their theoretical and practical aspects. The discussion includes a proof of variance reduction's ability to estimate expected values of response sums and response differences for paired functions within stochastic processes. Also discussed are the efficiencies of antithetic and random numbers used as variance reductions in varying problems with differing assumptions. The application of these variance reduction systems to reliability studies, network models and queueing systems is also examined.

Research, Stochastic analysis, Simulation methods, Simulation, Monte Carlo method, Monte Carlo methods

---

A Note on 'Efficiency of the Antithetic Variate Method for Simulating Stochastic Networks'

Article Abstract:

Previously a theorem was published that proved a negative covariance will be guaranteed. Thus a variance reduction is achieved with the application of antithetic sampling in stochastic networks where the arc times are symmetrically distributed about their means. This note proves that variance reduction is guaranteed for PERT type networks even when the arc time distribution is not symmetric.

Management science, Statistics (Data), Network Models, Stochastic Model, Statistics, Variance Reduction

Similar abstracts:

• Abstracts: A sensitivity analysis approach for some deterministic multi-criteria decision-making methods. Sensitivity analysis for dependent variables
• Abstracts: Deriving and analysing optimal strategies in Bayesian models of games
• Abstracts: Win/win business sales. A marketing strategy analysis of small retailers. The most critical problem for the fledgling small business: getting sales
• Abstracts: The paradox of "corporate culture": reconciling our selves to socialization. Assessing the Nestle boycott: corporate accountability and human rights
• Abstracts: A mixed integer programming model for solving a layout problem in the fashion industry. A cluster - analytic approach to facility layout

This website is not affiliated with document authors or copyright owners. This page is provided for informational purposes only. Unintentional errors are possible.