2017 q16 - Continuous Approximation Model
Hi. So, what is a continuous approximation model exactly? The term is never used in this lesson. Is this both single distribution models and collective models? Thanks.
Hi. So, what is a continuous approximation model exactly? The term is never used in this lesson. Is this both single distribution models and collective models? Thanks.
Comments
As you say, a continuous approximation model isn't really defined anywhere in the source material. The closest the source comes is mentioning that applying Panjer's algorithm is an example of a collective risk model.
In Panjer's algorithm we specify the frequency distribution (e.g. Poisson) and then combine it with a discretized severity distribution to produce an aggregate loss (pure premium) model known as a collective risk model. If we increase the number of points used in the discrete severity distribution then we can approximate any continuous severity distribution. When we do this, the collective risk model produced by Panjer's algorithm is also known as a continuous approximation model.
To be honest, I don't think you need to worry too much about this terminology beyond knowing the strengths and weaknesses of a collective risk model. If you have time, make sure you can apply Panjer's algorithm as it's in the source both here and in the NCCI circular but it's convoluted and there are probably other areas you should study first that are likely to get you more points on the exam.