Estimating the Parameter Risk of a Loss Ratio Distribution—Revisited
By Avraham Adler
When building statistical models to help estimate future results, actuaries need to be aware that not only is there uncertainty inherent in random events (process risk), there is also uncertainty inherent in using a finite sample to parameterize the models (parameter risk). This paper revisits Van Kampen (2003) in replicating its bootstrap method and compares it with measures of parameter uncertainty developed using maximum likelihood estimation and Bayesian MCMC analysis.
Keywords: Parameter risk, bootstrap, maximum likelihood, Bayesian MCMC, JAGS, Stan, R, approximate Bayesian computation