Introduction.w09

Introduction.w09 - ACTSC 433/833 Analysis of Survival Data...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: ACTSC 433/833 Analysis of Survival Data Johnny S.H. Li Department of Statistics and Actuarial Science, University of Waterloo January 6, 2009 Why ACTSC 433/833? About the Course About the Instructor A Required Course? I need ACTSC 433 to graduate! Why ACTSC 433/833? About the Course About the Instructor Relationship with Other Courses In ACTSC 232 and ACTSC 331: Life insurance Survival distributions and life tables Life annuities Benefit premiums Benefit reserves Why ACTSC 433/833? About the Course About the Instructor Relationship with Other Courses ACTSC 433/833 plays an important role in life insurance mathematics: Life insurance Survival distributions and life tables Life annuities ACTSC 433/833 Benefit premiums Benefit reserves Why ACTSC 433/833? About the Course About the Instructor Relationship with Other Courses Further relations: ACTSC 331 Life Contingencies ACTSC 453 Pension Mathematics ACTSC 433/833 ACTSC 431 Loss Models ACTSC 232 Actuarial Mathematics Why ACTSC 433/833? About the Course About the Instructor Relationship with SoA Exams ACTSC 433/833 covers part of the SoA Exam C syllabus. Specifically, it covers: 1 Part G: construction of empirical models; Part H: construction and selection of parametric models. 2 Why ACTSC 433/833? About the Course About the Instructor Some Facts about Exam C Number of Questions from Parts G and H 20 18 16 14 12 10 8 6 4 2 0 May00 9 14 11 10 15 14 18 18 19 Nov00 May01 Nov01 May02 Nov02 May03 Nov03 May04 Nov04 May05 Nov05 On average, 14.3 (out of 40) questions are based on parts G and H! This article provides an overview of the control cycle including a description of its key elements. Why ACTSC 433/833? About the Course An “optim al so lut ion” is, of co urse, th e goal will b e d efin ed fo r you b y a client o r b y yo ur About the Instructor of an y p roblem so lver. Wh en yo u b egin to manager. D id th ey get it right ? At oth er review th e p rob lem s set out in th is essay, fotimes a pro blem w ill b e ident ified b y yo u. cus on how con sist ently applying th e p rin ciDid yo u g et it right? In eith er case, you must ples of th e Cont rol Cycle1 will m ake th e ateffect ively commun icate and do cum ent th e tainment of an opt imal solutio n mo re likely. problem for all th e stakeh old ers. The comAs yo u consider th is small b ut rep resentat ive municatio n and do cum entat ion m ust id entify set of p rob lem s, t hin k about h ow each p robthe nat ure and potential impact of the risks lem co uld evolve on an ongoing basis; how th ey are present ed in th is module, how they are likely to b e expand ed in sub sequent modules and how they may affect yo ur career as an act uary. A quick review of the graph ic of the Contro l C ycle should make a few prelimin ary concept s evident to you. First, the actuary and h is p rofessionalism surround th e w ork. Second, ext ernal fo rces directly influence both the actuary and the actuary’s work. Th ird, while th ere are th ree d ist in ct stages in th e cont rol cycle, it is a cont inuous p rocess. In th e real wo rld, an act uary m ust manage th is continuous pro cess, anticipat e changes t o th e external fo rces and be aware of em erg ing p roblem s. necessary that yo u con stant ly con sider whether you are so lving th e right pro blem. Did yo u really get it right ? Have you ident ified and are yo u making pro vision for all th e material risks? Are th e risks yo u ident ified truly m aterial? Let ’s have no m illion do llar solutio ns to t en do llar p rob lem s. During the “Design th e So lution” stage yo u must communicate the problem w ith the parties, do cum ent your work and have your wo rk peer review ed b y oth er actuaries. If you reread th e previous sent en ces you will undoubted ly not e that the three item s m entioned could also b e thought of as “defin ing”, “design ing” and “monitoring”. Th e Cont rol Cycle not on ly describ es the “big picture” but it is also embedded thro ughout. The so lution design w ill vary g reat ly dep end ing on the t ype of p rob lem . It m ay be n ecessary to develop a model upon w hich t ests can be run. The model must incorp orat e th e abilit y to test the impact of each of th e critical risks. The assumptions used in build ing th e model m ust be selected carefully and must be con sistent wit h th e mat erial risks ident ified in the definit ion of th e p rob lem. The th ird stag e, “Mo nito r th e Results,” more than eith er of th e oth er two elem ents of th e Control Cycle, is what disting uish es an actu- ACTSC 433/833 is Practical As yo u review th e samp le problems think about wh at might be in vo lved in each stage of the Contro l C ycle. How w ould yo u “Define the Problem”? Som etim es the problem 1 “TheolvActuarial Control Cycle” inv ed. Before yo u p roceed to th e second stag e of the Cont rol Cycle (D esign th e So lut ion) it is imperative that all part ies ( e.g., you, yo ur client and yo ur manager) understand the problem and agree that it has b een defin ed ap- Why ACTSC 433/833? About the Course About the Instructor The Actuarial Control Cycle: An Example Observation: A term-life insurance product yields low profit. Define problem: 1 2 3 low sales? mortality rates are wrong? bad investment yield? Data analysis skills can help you identify the problem. Why ACTSC 433/833? About the Course About the Instructor The Actuarial Control Cycle: An Example Design solution: 1 2 Suppose that a wrong mortality assumption is the real problem. A possible solution: revise mortality tables. Data analysis skills can help you solve the problem. Why ACTSC 433/833? About the Course About the Instructor The Actuarial Control Cycle: An Example Monitor solution: 1 2 The solution may be wrong. Even if it is correct now, it may be wrong in the future. Data analysis skills can help you monitor the solution. Why ACTSC 433/833? About the Course About the Instructor In Addition to Exam C Materials... We will discuss some contemporary issues. In recent years, actuaries are facing the following problems: 1 2 survival distributions are changing; how survival distributions will change is uncertain. ◦ E.g., e 60 in 2005 is five years longer than was anticipated in the projections made in 1980s. Why ACTSC 433/833? About the Course About the Instructor The Actuaries’ Problem Longevity risk has become a hot topic in recent years. Venues for discussion of longevity risk: 1 2 Society of Actuaries “Living to 100 and Beyond” Symposium; International Longevity Risk and Capital Market Solutions Symposium. Why ACTSC 433/833? About the Course About the Instructor The Actuaries’ Problem The problem of longevity risk will be discussed in this course. We will also talk about possible solutions to the problem: 1 2 stochastic modeling; mortality derivatives. Examples of mortality derivatives: 1 2 longevity bond issued by BNP Paribas and EIB in 2004; QxX by Goldman Sachs in 2007. Why ACTSC 433/833? About the Course About the Instructor Teaching Plan Non-parametric estimation (G1,2,4) Parametric estimation (H1,2,3,4) Model selection and hypothesis testing (H5) Models for joint-life mortality (new in W09) Models with covariates (G3) Contemporary issues Weeks 1–4 Weeks 5–6 Weeks 7–8 Week 9 Weeks 10 Weeks 11–13 Why ACTSC 433/833? About the Course About the Instructor Reference Materials Lecture notes (printed and handwritten). Textbook: “Loss Models: From Data to Decisions” (3rd ed.; 2nd ed. is also acceptable). Journal articles and conference presentations. Why ACTSC 433/833? About the Course About the Instructor Grading ACTSC 433: Assignments Mid-term Final Exam 15% 20% 65% Note: You may see essay-type questions in assignments / mid-terms / the final exam. Why ACTSC 433/833? About the Course About the Instructor Grading ACTSC 833: Option 1: Assignments Final Exam Project Option 2: Project & presentation 100% 15% 65% 20% Why ACTSC 433/833? About the Course About the Instructor Grading ACTSC 833: Option 1: Assignments Final Exam Project Option 2: Project & presentation 100% 15% 65% 20% Why ACTSC 433/833? About the Course About the Instructor Your Instructor... Johnny Li: shli@uwaterloo.ca; MC6111; x35542 Joined faculty in July 2007 B.Sc., M.Phil (HKU); Ph.D. (Waterloo) Fellow of the Society of Actuaries (FSA) Why ACTSC 433/833? About the Course About the Instructor Your Instructor... Johnny Li: shli@uwaterloo.ca; MC6111; x35542 Joined faculty in July 2007 B.Sc., M.Phil (HKU); Ph.D. (Waterloo) Fellow of the Society of Actuaries (FSA) Why ACTSC 433/833? About the Course About the Instructor Your Instructor... Johnny Li: shli@uwaterloo.ca; MC6111; x35542 Joined faculty in July 2007 B.Sc., M.Phil (HKU); Ph.D. (Waterloo) Fellow of the Society of Actuaries (FSA) Why ACTSC 433/833? About the Course About the Instructor Your Instructor... Johnny Li: shli@uwaterloo.ca; MC6111; x35542 Joined faculty in July 2007 B.Sc., M.Phil (HKU); Ph.D. (Waterloo) Fellow of the Society of Actuaries (FSA) ...
View Full Document

This note was uploaded on 10/30/2009 for the course ACTSC 433 taught by Professor Johnny during the Winter '09 term at Waterloo.

Ask a homework question - tutors are online