1. Posts in the subject areas are now being moderated. Please do not post any details about your exam for at least 3 working days. You may not see your post appear for a day or two. See the 'Forum help' thread entitled 'Using forums during exam period' for further information. Wishing you the best of luck with your exams.
    Dismiss Notice

Query on incomplete triangle

Discussion in 'SP7' started by StartedThinking, Mar 19, 2018.

  1. StartedThinking

    StartedThinking Keen member

    Hi,
    I have a query on incomplete triangle (latest diagonal being incomplete).
    Let us suppose that we have a cumulative incurred triangle (annual triangle) where the latest diagonal has only 8 months data in it. I thought to apply 8-12, 20-24, 32-36 month development factors respectively for the latest diagonal from the previous years data, which gives me an yearly triangle - now being used as input triangle, for which I want to apply Chain ladder method for Ultimate.

    But my friend says that we should use 8-20, 20-32, 32-44 etc development factors from the previous year triangles and develop them to ultimate, But I am not convinced with it. Can you please help me understand which is the better methodology to use --- from reserving viewpoint and if possible from pricing viewpoint.

    Thanks in advance.
     
    StartedThinking, Mar 19, 2018
    #1
    Glykeria likes this.
  2. Darren Michaels

    Darren Michaels ActEd Tutor Staff Member

    Probably the most straight forward / common thing to do would be to "gross-up" the latest diagonal in some appropriate way (eg increase pro-rata) to make it annual as per the other diagonals and then apply the chain ladder method as normal to annual triangles.

    As ever, what you might do in practice would depend on the specifics of the class concerned, eg if seasonality was a big issue a simple pro-rata adjustment would unlikely to be appropriate.
     
    Darren Michaels, Mar 21, 2018
    #2
    Glykeria likes this.

Share This Page