2017 Introduction to Applied Statistics Limited Attendance Seminar

Event Details

-

Marriott Château Champlain
1050 Rue de la Gauchetière O
Montréal, QC H3B 4C9
Canada

March 13, 2017

About This Event

Applied statistical knowledge is important to Casualty Actuaries in their core areas of practice: reserving and ratemaking. Predictive modeling is also a tool that is increasingly being embraced by insurance companies, and knowledge of applied statistics is a foundation for understanding predictive models.

Event Information

Applied statistical knowledge is important to Casualty Actuaries in their core areas of practice: reserving and ratemaking. Predictive modeling is also a tool that is increasingly being embraced by insurance companies, and knowledge of applied statistics is a foundation for understanding predictive models.

In recognition of the need to educate actuaries on predictive modeling, the CAS and Canadian Institute of Actuaries sponsored a book on predictive modeling that focuses on insurance applications and examples. Because we view the book as a tool for analytics for actuaries, we recommend it as a resource for students attending the course. We recommend that participants read chapter 2 “Overview of Linear Models” of Volume 1 of the book "Predictive Modeling Applications in Actuarial Science".

In addition, we will incorporate concepts from a Facilitators Guide to “Testing Assumptions Underlying Estimates of Loss Reserves” that was developed as a project of the Committee on the Theory of Risk. The tests in the guide and in the paper that provided the motivation for the project are primarily regression based and can be applied using Microsoft Excel. We will supply several data sets relevant to common actuarial applications to use for the in-class exercises.

This course will cover topics needed to use, understand and apply basic analytics modeling methods with an emphasis on linear models and regression as well as an introduction to generalized linear models (GLMs). This course will stress a hands-on approach. The software required to do the homework and exercises will be Microsoft Excel and R. We recommend that students download and install R. As an editor for R, we recommend RStudio but Notepad and Word will work. Advanced knowledge and experience in R will not be required. The first hour in the morning of the first day will provide a brief overview of all the topics to be covered along with a more detailed review of some introductory topics.

OBJECTIVE

The seminar will introduce applied statistics to actuaries so that attendees will understand the statistical underpinnings of commonly used actuarial methods, such as the chain ladder reserving method and common approaches used in ratemaking. Also participants will have the ability to use applied statistical models in Excel and R. The course focus includes applications to reserving, ratemaking and claims databases. This course will be appropriate for both actuarial students and more experienced actuaries. However, it will assume that participants are familiar with basic reserving and ratemaking methods.

At the course’s conclusion, the students should have gained a general understanding of statistics used in common actuarial applications, along with the theory underlying the use of specific statistics. The participants should be able to interpret the statistical work of other actuaries and statisticians and be able to apply the methods introduced in the course, particularly regression and logistic regression, in their own work.

CONTINUING EDUCATION CREDIT

The CAS Continuing Education Policy applies to all ACAS and FCAS members who provide actuarial services. Actuarial services are defined in the CAS Code of Professional Conduct as “professional services provided to a Principal by an individual acting in the capacity of an actuary. Such services include the rendering of advice, recommendations, findings, or opinions based upon actuarial considerations.”

Members who are or could be subject to the continuing education requirements of a national actuarial organization can meet the requirements of the CAS Continuing Education Policy by satisfying the continuing education requirements established by a national actuarial organization recognized by the Policy.

Participants should claim credit commensurate with the extent of their participation in the activity. CAS members earn 1 CE Credit per 50 minutes of educational session time not to include breaks or meals.

Note: The amount of CE credit that can be earned for participating in this activity must be assessed by the individual attendee. It also may be different for individuals who are subject to the requirements of organizations other than the American Academy of Actuaries.

CONTACT INFORMATION

  • For more information on content, please contact Nora Potter, Education Coordinator, at npotter@casact.org.
  • For more information on attendee registration, please email the Actuaries' Resource Center at arc@casact.org.
  • For more information on the Seminar other than registration or content issues, please email meetings@casact.org.
  • For more information on other CAS opportunities or regarding administrative policies such as complaints and refunds, please contact the CAS Office at (703) 276-3100 or office@casact.org.

 

Event registration

REGISTRATION FEES

Print Registration is also available for download.

REGISTRATION FEES (IN U.S. DOLLARS)

IF RECEIVED ON/BEFORE MARCH 13

IF RECEIVED
AFTER MARCH 13

CAS Member, or Active Candidate*

$875

$975

Non-Members

$1,075

$1,175

* An Active Candidate is a non-CAS member who has attempted at least one actuarial exam in the last two years.

Cancellation Policy

The registration fee will be refunded for a cancellation received on or before March 30, 2017, less a $100 administrative fee. Only written cancellations will be honored. Cancellation requests can be faxed to (703) 276-3108 and e-mailed to refund@casact.org.

 

Lodging

LODGING INFORMATION

Hotel Room Rate and Reservations
Marriott Château Champlain
1050 Rue de la Gauchetière O
Montréal, QC H3B 4C9
Canada

The room rate is $189 (CND) for single or double occupancy, plus applicable taxes. Reservations must be made prior to March 13, 2017.

To place a reservation, call TBD and noting the Casualty Actuarial Society room block.

While reservations must be made prior to March 13, 2017, in order to receive the $189 (US) rate, there is no guarantee that rooms will be available should you wait until this date. CAS strongly suggests that you make your room reservations early to receive a room before the room block fills.

English: Book your group rate for Applied Statistics LAS
French: Réservez votre tarif pour Applied Statistics LAS

Schedule

SCHEDULE

Day 1: Introduction to Statistics and Refresher on Linear Models - 8:30 AM - 5:00 PM

  • Introduction to course
    • Objectives
    • Introduction to Predictive Modeling Applications in Actuarial Science book
    • Other resources
    • Introduction to datasets used in course
      • Mack loss development triangle
      • California Auto Assigned Risk Data (CAARP)
      • Fraud (Questionable claims data)
    • Illustration of models that can be applied to datasets including, linear regression, ANOVA and contingency tables
    • How to download the open source software R
    • How to download and install libraries for R
    • Resources for R
  • Data Exploration
    • Eyeball data
    • Examination of summary and descriptive statistics
    • Box and whisker plots
    • Scatterplots
    • How are outliers addressed?
    • Application to CAARP Data
  • Use of Excel’s regression functions
  • Interpreting the outputs from a linear model and hypothesis testing
  • Application to reserving: chain ladder as regression through the origin
  • Application to rate making – estimating a trend
  • Exercise using regression
    • Applications to loss development data sets
  • Hypothesis testing versus holdout sample and cross-validation
  • Alternatives to linear model
  • Transformation of variables
  • Heteroscedasticity
  • Nonlinear regression
  • Binning continuous variables
  • Categorical predictor variables and dummy variables
  • Exercise: predict pure premium
  • Exercise: Test assumptions of chain ladder
  • Binary dependent variables
  • Cross classification and contingency tables
  • Regression with Categorical dependent variables
    • Binary regression
    • Logistic regression
    • Application to questionable claims data

Day 2: Generalized Linear Models - 8:30 AM - 12:00 PM

  • Shortcomings of the linear model: non-normality of data, discrete outcomes, nonlinearity, etc.
  • Exponential family of distributions
  • Link functions
  • Form of generalized linear model
  • Importance of variance when fitting generalized linear model
  • Examples including modeling frequency, severity, and probability of an event. R will be used to illustrate the examples
  • Fitting a glm: maximum likelihood estimation and reweighted least squares
  • Quasilikelihood estimation
  • Interpreting statistics in a GLM: deviance, AIC/AICC/BIC, chi-squared, loglikelihood
    • Exercise: estimating frequency
    • Exercise: estimating severity
    • Exercise: estimating questionable claims from fraud data

 

Schedule

SCHEDULE

Day 1: Introduction to Statistics and Refresher on Linear Models - 8:30 AM - 5:00 PM

  • Introduction to course
    • Objectives
    • Introduction to Predictive Modeling Applications in Actuarial Science book
    • Other resources
    • Introduction to datasets used in course
      • Mack loss development triangle
      • California Auto Assigned Risk Data (CAARP)
      • Fraud (Questionable claims data)
    • Illustration of models that can be applied to datasets including, linear regression, ANOVA and contingency tables
    • How to download the open source software R
    • How to download and install libraries for R
    • Resources for R
  • Data Exploration
    • Eyeball data
    • Examination of summary and descriptive statistics
    • Box and whisker plots
    • Scatterplots
    • How are outliers addressed?
    • Application to CAARP Data
  • Use of Excel’s regression functions
  • Interpreting the outputs from a linear model and hypothesis testing
  • Application to reserving: chain ladder as regression through the origin
  • Application to rate making – estimating a trend
  • Exercise using regression
    • Applications to loss development data sets
  • Hypothesis testing versus holdout sample and cross-validation
  • Alternatives to linear model
  • Transformation of variables
  • Heteroscedasticity
  • Nonlinear regression
  • Binning continuous variables
  • Categorical predictor variables and dummy variables
  • Exercise: predict pure premium
  • Exercise: Test assumptions of chain ladder
  • Binary dependent variables
  • Cross classification and contingency tables
  • Regression with Categorical dependent variables
    • Binary regression
    • Logistic regression
    • Application to questionable claims data

Day 2: Generalized Linear Models - 8:30 AM - 12:00 PM

  • Shortcomings of the linear model: non-normality of data, discrete outcomes, nonlinearity, etc.
  • Exponential family of distributions
  • Link functions
  • Form of generalized linear model
  • Importance of variance when fitting generalized linear model
  • Examples including modeling frequency, severity, and probability of an event. R will be used to illustrate the examples
  • Fitting a glm: maximum likelihood estimation and reweighted least squares
  • Quasilikelihood estimation
  • Interpreting statistics in a GLM: deviance, AIC/AICC/BIC, chi-squared, loglikelihood
    • Exercise: estimating frequency
    • Exercise: estimating severity
    • Exercise: estimating questionable claims from fraud data