|
|
|
25 Years Ago in The Actuarial Review
by Walter C. Wright
The following is an extract from an article that discussed a presentation that a Dr. Freifelder gave at the 1977 Spring Meeting of the Casualty Actuaries of Philadelphia (now the Casualty Actuaries of the Mid-Atlantic Region). Aren't these issues still being debated on CASNET?
Dr. Freifelder compared utility theory as a ratemaking technique to mean-variance theory and ruin theory, which currently underlie many non-life ratemaking procedures.
In his discussion of risk, Dr. Freifelder noted that both the mean-variance and ruin theory techniques are developed from the principles of classical statistics and probability theory. The former suggests the use of the variance or standard deviation as a risk measure, while the latter method measures risk by the probability of large losses. Only utility theory measures risk by looking at the entire probability distribution….
The use of an exponential utility theory premium calculation principle changes the problem of insurance ratemaking into a problem of determining the probability distributions that best fit the empirical loss data on frequency and severity. Because a large amount of data is available, the problem of specifying the proper distributions should not be too difficult.