Estimating class relativities - Finger vs Mclenahan

[Purple Princess]

Finger and Mclenahan are seriously messing with my mind.

I did 2004 #36 and got it wrong because I didnt divide the on-level earned premium by the current class relativity before calculating the loss ratios. So then I was like, I won't make that mistake again!!!

So I get to 2005 #49, and first I make the earned premium on-level by dividing by year's base rate and multiplying by 2005 base rate. Then, because I learned my lesson in 2004 #36, I divide the on-level earned premium by the current relativities. But according to the solution, this is wrong!!! I am soooooo confused.

I did a little research and it turns out that the solution to 2004 #36 is the method shown in Mclenahan (p. 119) and the solution to 2005 #49 is the method shown in Finger (p. 325-326). And since they don't tend to mention which article to use, there is no way for us to know which method to use.


So do you guys think that we could get credit for either method on the exam? I am assuming I could just state which author's method I was using. Right?

[2M]

They should give you the same result.

Finger:

off - balance = (sum of premium * relativities)/sum of premium

off balance goes in the denominator to get your adjusted rate

McClenahan:

off - balance = sum of premium/(sum of premium * relativities)

this is a direct multiplicative after to get your adjusted rates

[hzhang]

PP: I agree with you that Finger does not take current relativity into account, while McClenahan does when they are both calculating the indicated relativities. However, from the layout of the problems, you could tell immediately the problem is from Finger's exhibit or McClenahan's one. That's how I approach that type of problems.

[Purple Princess]

Originally Posted by hzhang

PP: I agree with you that Finger does not take current relativity into account, while McClenahan does when they are both calculating the indicated relativities. However, from the layout of the problems, you could tell immediately the problem is from Finger's exhibit or McClenahan's one. That's how I approach that type of problems.

The layout of 2005 #49 closely resembles a Mclenahan exhibit because the first column is current relativities, and yet the answer uses Finger's method.

[pcact]

Notice that:

EP @ 2005 is not used to calculate indicated relativities.

The base rates of A and B are different and independent of the current relativities. (Can this happen?)

[hzhang]

Originally Posted by Purple Princess

The layout of 2005 #49 closely resembles a Mclenahan exhibit because the first column is current relativities, and yet the answer uses Finger's method.

The first column of current relativities is to "mislead" you. Look at the rest columns. Finger's exhibit has experience period and credibility information, while Mclenahan's one does not.

[pcact]

Originally Posted by Purple Princess

The layout of 2005 #49 closely resembles a Mclenahan exhibit because the first column is current relativities, and yet the answer uses Finger's method.

You can use McClenahan's method but it produces a different answer than Finger's.

[hzhang]

Originally Posted by pcact

Notice that:

EP @ 2005 is not used to calculate indicated relativities.

You're right. When we are asked to calculated indicated relativities, we do NOT need to calculate off-balnce which is based on EP @ 2005. However, if we're asked to calculate indicated base rate, we need that information.

Originally Posted by pcact

The base rates of A and B are different and independent of the current relativities. (Can this happen?)

It shows in this problem that they are independent and not consistent.

[hzhang]

Originally Posted by pcact

You can use McClenahan's method but it produces a different answer than Finger's.

How do you do to incorporate experience information and credibility in McClenahan's method? -- Is that out of scope of the syllabus? We're NOT talking about off-balance.

[pcact]

Originally Posted by hzhang

It shows in this problem that they are independent and not consistent.

In Finger's paper p.322, current relativity of B = current base rate of B / current base rate of A.