Scribd Logo
Upload

Welcome to Scribd!

  • 83 views

    Term Test 2016 Act466

    Uploaded by

    ZenitharX
    This document appears to be instructions for a term test in ACT466 at the University of Toronto. It provides information about the structure and rules of the exam, including: - It is a closed book test lasting 90 minutes - Only non-programmable calculators are allowed - Students must stay seated and not talk until papers have been collected - Photo ID must be on the desk during the exam - The test contains 6 questions worth 10 points each about insurance concepts like claim distributions, credibility standards, and Bayesian analysis.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Term Test 2016 Act466

    Uploaded by

    ZenitharX
    83 views3 pages
    This document appears to be instructions for a term test in ACT466 at the University of Toronto. It provides information about the structure and rules of the exam, including: - It is a closed book test lasting 90 minutes - Only non-programmable calculators are allowed - Students must stay seated and not talk until papers have been collected - Photo ID must be on the desk during the exam - The test contains 6 questions worth 10 points each about insurance concepts like claim distributions, credibility standards, and Bayesian analysis.

    Original Description:

    ACT 466 U of T

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document appears to be instructions for a term test in ACT466 at the University of Toronto. It provides information about the structure and rules of the exam, including: - It is a closed book test lasting 90 minutes - Only non-programmable calculators are allowed - Students must stay seated and not talk until papers have been collected - Photo ID must be on the desk during the exam - The test contains 6 questions worth 10 points each about insurance concepts like claim distributions, credibility standards, and Bayesian analysis.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    83 views3 pages

    Term Test 2016 Act466

    Uploaded by

    ZenitharX
    This document appears to be instructions for a term test in ACT466 at the University of Toronto. It provides information about the structure and rules of the exam, including: - It is a closed book test lasting 90 minutes - Only non-programmable calculators are allowed - Students must stay seated and not talk until papers have been collected - Photo ID must be on the desk during the exam - The test contains 6 questions worth 10 points each about insurance concepts like claim distributions, credibility standards, and Bayesian analysis.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 3

    NAME: ———————————————————————————

    STUDENT I.D.: ———————————————————————–

    University of Toronto
    ACT466 - TERM TEST
    25th of February 2016
    Instructor - Andrei Badescu

    NOTES:

    1. Only non-programmable calculators are allowed.

    2. A booklet is to be handed in with this exam paper.

    3. This is a closed book test.

    4. Timing: 90 minutes.

    5. Please stay in your seats and do not talk till all papers have been collected.

    10. Photo ID on desk during exam please.

    11. Good luck!

    1
    1. (10 points) You are given the following information:

    i) the number of claims follows a Poisson distribution.

    ii) Claim severity is independent of the number of claims and has the following distribution,
    7
    f (x) = 52 x− 2 , x > 1.

    a) (3 points) Determine the coefficient of variation of the claim severity distribution.

    b) (7 points) A full credibility standard is determined so that the total number of claims is within
    5% of the expected number with probability 98%. If the same exmpected number of claims for
    full credibility is applied to the total cost of claims, the actual total cost would be within 100K%
    of the expected cost with 95% probability. Using the normal approximation of the aggregate loss
    distribution, determine K.

    2. (10 points) For an insurance portfolio, you are given the following:

    i) The number of claims for each insured follows a Poisson distribution.

    ii) The mean claim count for each insured varies. The distribution of the mean claim counts is a
    gamma distribution with α = 0.5 and θ = 4.

    iii) The size of the claims for each insured follows a Pareto distribution with parameters α = 3
    and θ = 6000.

    iv) The credibility standard is that the aggregate claims must be within 10% of expected p of the
    time

    v) 1812 claims were observed, resulting in 80% credibility.

    Find p.

    3. (10 points) There are two classes of insureds, A and B.

    For class A, claim frequency follows a distribution which is a mixture of a Poisson with mean 0.1
    with weight 2/3 and a Poisson with mean 0.2 with weight 1/3.

    For class B, claim frequency follows a distribution which is a mixture of a Poisson with mean 0.1
    with weight 1/3 and a Poisson with mean 0.2 with weight 2/3.

    Half of the insureds are in each class. A certain insured submits at least one claim in the first year.
    Calculate the probability that the same insured submits at least one claim in the second year.

    2
    4. (10 points) You are given the following probability distribution for the prior hypothesis Θ:

    Θ
    0 1
    0 0.4 0.1
    X 1 0.1 0.2
    2 0.1 0.1

    Two observations for a randomly selected individual are 0, 1. Calculate the Bayesian prediction
    of the expected value of the next observation from this individual.

    5. (10 points) Claim sizes follow an inverse Pareto distribution with parameters τ = 2 and θ. The
    improper (does not integrate to 1) prior density of θ is π(θ) = 1θ , θ > 2. You observe a claim of size
    1. Determine the posterior expected value of θ.

    6. (10 points) Claim frequency follows a mixture of a Poisson distribution with mean 0.1 with
    a weight α and a Poisson distribution with mean 0.2 with a weight 1 − α. The weight α varies by
    insured, and has a probability density function f (α) = 2(1 − α), 0 < α < 1. There is one claim in
    year 1. Determine the posterior mean of α.

    You might also like