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

Buhlmann-Straub

Discussion in 'SP8' started by Nicholas.Campbell, Feb 25, 2018.

  1. Nicholas.Campbell

    Nicholas.Campbell Member

    Silly question(s) but...

    Is the Buhlmann-Straub model and EBCT 2 the same model?

    Also the credibility factor z = n/n+k, where k = E[s^2(theta)] / var[m(theta)] on page 29 is referring to EBCT 1?

    And when it says on page 29, "Note the similarities to this approach (Buhlmann-Straub) and the Bayesian credibility approach", it's referring the fact that the estimate can be decomposed into a weighted average of the observation and the benchmark? (I initially thought the Buhlmann-Straub and EBCT were two different models and the notes were pointing out how similar they were!)
     
    Nicholas.Campbell, Feb 25, 2018
    #1
  2. Ian Senator

    Ian Senator ActEd Tutor Staff Member

    Not a silly question. They're not identical, but are incredibly similar. There may be slight differences in the assumptions, but we've tried to pin down the differences and it gets messy very quickly. It's made all the more complicated by the fact that CT6 and ST8 use different parameter names and terminology.
    My advice is just concentrate on the ST8 assumptions and method, ignore the EBCT stuff from CT6, try to avoid looking too hard at pages 29/30 in the Tables, and just do what you need for ST8. Also bear in mind that the BS method came up in the last exam...!
     
    Ian Senator, Mar 13, 2018
    #2
    Hemant Rupani likes this.
  3. Nicholas.Campbell

    Nicholas.Campbell Member

    Thanks Ian - yes, they are incredibly similar, but i will stick to ST8.
     
    Nicholas.Campbell, Apr 1, 2018
    #3

Share This Page