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

Short term reserves calculation

Discussion in 'SP1' started by Trevor, Sep 17, 2021.

  1. Trevor

    Trevor Ton up Member

    Hi, I want to understand the links between different "categories" or calculation approach.

    Throughout past paper practices, I've seen a few "categories" of calculation approaches
    1. Case-by-case estimates/statistical approach
      In this case, we are expected to say which approach to use, and then maybe explain how to carry out this approach
      Example questions: 2014 September Q3

    2. Actually describing the formula/model
      In this case, we are expected to describe the model/formula to calculate reserves rather than which approach
      Examples: 2013 September Q5iii, 2020 April Q5ii

    Comparing these few sets of question, the questions mean the same thing to me: "How to calculate reserves"
    When faced this question, I have to guess what the examiner wants me to say.




    In general, I would like to understand their links (Please correct me if I am wrong)
    When we want to calculate reserves, we consider the steps in two levels:
    1. Do we want to make a rough guess (case estimate) or actually refer to statistical data (statistical estimate)?
    2. If we choose the latter, and then we decide the formula/model to use. "statistical estimate" will mean setting reserve assumptions based on past data.

    Can anyone explain the relationship between these 2 "categories" of answers? I am really worried to face reserve question in exams, not knowing which one scores mark.
     
    Trevor, Sep 17, 2021
    #1
  2. Mark Willder

    Mark Willder ActEd Tutor Staff Member

    Hi Trevor

    I'll take your two steps first of all. Yes, we need to make a choice between case estimates and statistical estimates when calculated the outstanding claims reserve. I wouldn't describe case estimates as being rough estimates - these are reserves set by highly trained staff by looking at all the information on the case at hand. In fact I would describe statistical estimates as being the rough estimates - we completely ignore any specific information about the individual claim and hope that on average the cost is similar to past claims - sometimes this rough statistical estimate will turn out to be far too much or far too little, but when we average over hundreds of claims it probably isn't too bad. Yes, statistical estimates mean setting reserves based on data from past claims.

    Now looking at the reserving questions you mentioned from past exams. Yes, these questions all ask how to calculate reserves, but the questions are worded very differently making it clear that the answers will be very different. September 2013 Q5 asks about the reserves for group IP - we have no specific information about what stage the policy is at, so we must give the full range of reserves: UPR, IBNR, outstanding claims reserve etc. In contrast, in September 2014 Q3 we are told that the first policy is a large cancer claim, so we need only talk about setting the outstanding claims reserve and there will be no marks for UPR and IBNR - to ensure that we have different answers to each part, we'll need to draw out the differences in the claims that determine whether case or statistical estimate are required.

    Best wishes

    Mark
     
    Mark Willder, Sep 20, 2021
    #2
    Trevor likes this.
  3. Trevor

    Trevor Ton up Member

    Hi Mark,

    This is very clear now, I was trying to analyse the command verb, being “set” or “methodology” although they mean pretty much the same thing but haven’t realize the entire question scenario.

    My understanding now is that, in general, we will use (answer with) the formula/model approach if asked for a book of policies. Unless we are given a particular large/unusual policy like in the 2014 paper, which warrants a special treatment for that one policy out of the book; therefore selecting between case/statistical estimate.

    Thanks again Mark.
     
    Trevor, Sep 20, 2021
    #3

Share This Page