Brosius Assmuptions

[Ballin]

In Brosius it states that the underlying assumption of the least squares method is "that year to year changes in loss and loss reporting distributions are small or can be corrected for" (page 14)

It also states that the least squares method "is worth considering whenever random year to year fluctuations in loss experience are significant." (page 1)

If on an exam they asked you when to use the L-S method. What would you say? These statements lead to very different answers

Thanks for your input/thoughts!

[booyah81]

There are a ton of past exam question on this. I think each time what they're looking for is that least-squares is only appropriate if year-to-year fluctuations are due to random changes, and may not be appropriate if year-to-year changes are due to systematic changes.

[BurrNuts]

I think the key with the first one is the word "distribution." The distribution can have a lot of variance which could result in large swings in "experience" from year to year. So long as the large swings are due to volatility of the experience rather than changes in the underlying distribution, then the assumption is met.