Policy Year data trend period

[jk]

I'm confused by an apparent contradiction in how to trend policy year data between the Feldblum (WC) article and the Graves/Castillo (CGL) article.

Feldblum says that you trend from the average accident date of the experience period. If you have PY 2002 data, for example, this covers policies written between 1/1/2002 and 12/31/2002 and expiring between 1/1/2003 and 12/31/2003. The average of the experience period is 1/1/2003. This makes perfect sense to me.

But Graves and Castillo say (page 661) that you trend PY data from the "average date of coverage", which appears to be a year before the end of the policy year. They trend PY 1986 data from 12/31/85. This carries over into exam problems such as 2003 #28.

Can anyone explain why they do it differently?

[Utanapishtim]

It's been a year now, but I believe it's an issue of the definition of policy year. If I recall correctly, the Graves & Castillo paper is the one that labels their policy years as 'ending 31/12/1986', which in fact refers to the policy year beginning 1/1/1985.

I could be wrong, but I remember that being an issue, at least.

Cheers.

[jk]

If I recall correctly, the Graves & Castillo paper is the one that labels their policy years as 'ending 31/12/1986', which in fact refers to the policy year beginning 1/1/1985.That's exactly right. I was wondering if that was an issue. Feldblum just says "Policy Year 1985", which I interpret as above. G/C say "Policy Year ending 12/31/1985", which I took (apparently incorrectly) to be synonymous. Am I to understand, then, that "Policy year ending 12/31/1985" indicates policies written during 1984 and expiring during 1985? Seems like an odd phraseology.

[MountainHawk]

That's exactly right. I was wondering if that was an issue. Feldblum just says "Policy Year 1985", which I interpret as above. G/C say "Policy Year ending 12/31/1985", which I took (apparently incorrectly) to be synonymous. Am I to understand, then, that "Policy year ending 12/31/1985" indicates policies written during 1984 and expiring during 1985? Seems like an odd phraseology.

It's been a few years, but I remember this aggravating the hell out of me too. I believe Ut... is right tho. It's terrible phraseology, but it's the only way the rest make sense.

[BassFreq]

Thank you, thank you, thank you, thank you!!!

Graves and Castillo is next on my reading list. This saves me some time and aggravation. I know I would have missed that subtlety.

[ebradford]

Ah yes.... the is one of the few things I love about being behind in my studying. Thanks guys.

-E