I would like to know what we would get when we subtract the last diagonal entries from ultimate when the we have Accident years cohorts and intimation development periods? How does this quantity compare with difference between ultimate and last diagonal entries of triangle with Accident year cohort and transaction development periods. Can someone help? Thanks a ton in advance for the help.
So the ultimate is the final amount paid for the premium earned for each accident period. The paid or incurred is the amount paid or reported up to a defined period in time I guess. The amount that is outstanding is the difference between these quantities. The last triangle is the ultimate so the difference is between the ult and that paid or reported to date. A first go at me for explanation so please correct me if I'm not clear....
Assuming 'intimation developments' means the number of [months] since the claim was reported. Then this is not useful or meaningful. For example let's say we have a AY/Dev Month triangle in November, then we get a new claim in december. This new claim is on it's first month so it will increase the value of the 2018/D1 cell (top left cell). So the most right diagonal does not have the latest value of claims. Off the top of my head, there will never be a time when you will want a triangle with different origin and development basis (there might be weird fringe cases but you shouldn't ever see it in exams or real life).
But I was told that this approach foreshortens the development of IBNER. Are you able to visualise this? Can one of the ActEd tutors help me out? Thanks for the time!
Hi Adithyan We can't really comment unless you give us more detail of what you're asking - I don't know what you mean by 'intimation' and I'm not sure what figures you're tabulating within the triangle, eg paid or incurred, in that year or cumulative, and if incurred then what does incurred include etc. Once we know these things, we can work out what the difference is between the ultimate and the latest diagonal and that will be your answer. Ian
Dear Sir, Thanks for your response. Let's assume the triangle to be incurred. Rows are accident quarter cohorts and columns are reported development quarters.
In that case it sounds just like a normal chain ladder triangle as covered in Subject CT6 (in old subjects). Doesn't really make much odds whether you're talking in quarters or years. So, assuming you've used cumulative data, your latest diagonal will be paid to date plus outstanding reported up to that point in time. The projection for each cohort will be the ultimate paid. So subtract the paid to date and o/s reported and what do you have left?
Correct! All that I wanted to understand is why is IBNER development foreshortened when the columns are development reported quarters while I get IBNR and adequate IBNER when development periods are transaction periods?
Sorry, still not entirely sure what you mean by transaction periods and development reported periods. If you can point me in the direction of where you're reading this - eg is it a past exam question, or something in the notes - then I can hopefully see what's trying to be said. Thanks Ian
