ParallelogramOLF - policy length

[andyhufford-HARC]

Is there documentation for the ParallelogramOLF the takes into account the length of the policy term (i.e. semi-annual vs. annual)?

approximation_grain was not described in the API Reference, but that appears to be monthly or daily grain noted from another forum (#497). Is there another parameter for policy-term length that is an input?

class chainladder.ParallelogramOLF(rate_history=None, change_col='', date_col='', approximation_grain='M', vertical_line=False)

[kennethshsu]

Hi @andyhufford-HARC!

I can (should) probably ad the approximation_grain to the doc site. This is not good that it's missing, so thanks for catching that.

And currently, I think the policies are assumed annual, but I think it'll be great to incorporate other lengths as well. This will be a great feature. Would you be interested in helping implement this? Or, if you want a smaller task, I think coming up with example calculations in Excel will be good too, so we can reconcile.

[andyhufford-HARC]

Hi @kennethshsu,

Thanks for the quick response and for opening the new issue (#686)! I’d definitely be interested in helping out with this. It’s a common real-world scenario we run into, so I think it would be a great addition to the library.

Full disclosure: I’m very much an actuary first and a coder second—our org is starting to implement chainladder operationally (currently we are spreadsheet driven), but I’ve never actually implemented anything or contributed on GitHub before.

I’d be happy to start by putting together some example calculations in Excel to reconcile against, as you suggested. That feels like a safe way for me to contribute while I learn the ropes of the repo. If you can point me toward the best place to share those or any specific format you prefer, I’ll get to work on them!

Looking forward to collaborating,
Andy

[kennethshsu]

That's great, do you mind putting a comment (anything) on #686 so we can claim that ticket?

[kennethshsu]

I opened a new issue here: #686