223020 (M22) An introduction to Markov chain Monte Carlo methods and their actuarial applications : Scollnik D.P.M.,: Casualty Actuarial Society, Proceedings, Volume LXXXIII, Part 1, Nr. 158, 1996, pp 114–165
- Publisher
- Elsevier Science
- Year
- 1998
- Tongue
- English
- Weight
- 91 KB
- Volume
- 22
- Category
- Article
- ISSN
- 0167-6687
No coin nor oath required. For personal study only.
✦ Synopsis
In making all-or-none choices between alternative securities, Samuelson (1997b) suggested that investors of different risk-aversion should calculate from past samples of those securities their relevant Harmonic Means, or Geometric Means, or other associative means representative of their respective degrees of relative-riskaversion. Here it is shown how this learning procedure can be improved upon when you have prior knowledge that the securities have log-normal distributions. Classical estimation theory, concerning consistent, efficient and sufficient statistics, is shown to have a cash value by means of the calculable measure of (ex ante) "riskcorrected certainty equivalents". Needed qualifications and testings are also presented.