P&C Financial Reporting Committee
Source:
2006 Regional Affiliate - OCCA
Type:
affiliate
Panelists:
Claudette Cantin
Keywords:
financial reporting, property/casualty
Catastrophe Models - Science, Speculation or Sinister?
Source:
2006 Regional Affiliate - OCCA
Type:
affiliate
Keywords:
catastrophe models
2006 CAS Ballot Issues
Source:
2006 Regional Affiliate - OCCA
Type:
affiliate
Panelists:
Jim Christie
Keywords:
ballot issues
Business Session
Source:
2006 Regional Affiliate - OCCA
Type:
affiliate
Keywords:
business session
CPD Presentation
Source:
2006 Regional Affiliate - OCCA
Type:
affiliate
Keywords:
CPD
Enterprise Risk Management
Source:
2006 Regional Affiliate - OCCA
Type:
affiliate
Panelists:
Pierre Laurin
Keywords:
ERM, Enterprise Risk Management
Actuarial Issues at OSFI
Source:
2006 Regional Affiliate - OCCA
Type:
affiliate
Panelists:
Dave Oakden
Keywords:
OSFI
Automobile Insurance Rate Filing Process
Source:
2006 Regional Affiliate - OCCA
Type:
affiliate
Keywords:
automobile insurance, rate filing
Business Sessions
Source:
2006 Regional Affiliate - OCCA
Type:
affiliate
Keywords:
business session
Precision Rating-Rating by Address Rather than Territory
Geographic location is an important differentiator of insurance risk. Traditionally, loss and exposure data are aggregated at the zip code level (or some other geographic unit) for the purpose of determining a measure of geographic risk for the zip code. Zip codes with similar loss experience are grouped together to form territories and territorial relativities. Because zip codes are not developed for insurance purposes, the zip codes include risks that are heterogeneous with respect to geographic insurance risk. This heterogeneity can cause inequitable rates and rate discontinuities at the boundaries. Zip codes-and thus boundaries based on zip codes--are further problematic as the US Postal Service can change them frequently.
Given today's technology and data, it may be possible to incorporate geography at a more granular level (e.g., address) and avoid these issues. Of course, to do this it becomes critical to supplement internal data with external data and to use different statistical techniques to more accurately determine the geographic risk. This session with explore the advances possible in precision geographic rating given today's technology, to include a discussion of practical issues such as gaining regulatory approval and implementing the more complex rating schemes.
Source:
2006 Spring Meeting
Type:
concurrent
Moderators:
Brian Clancy
Panelists:
Stephen Fiete, Daniel Finnegan
D&O Liability: What Happened and What's Ahead
Did premiums decline again in 2005? Did frequency and severity continue their upward trend? Which board members are interested in their D&O coverage and what are they doing about it? Findings from the recently released Tillinghast 2005 D&O Survey will be discussed along with the panel's view on where the D&O market is headed in 2006.
Source:
2006 Spring Meeting
Type:
Concurrent
Moderators:
Gerald Kirschner
Panelists:
Elissa Sirovatka, Vagif Amstislavskiy
From Class to Individual Rating
In this session, panelists will examine methods of using GLM-based technologies to bring rating down from broad classes of risk to estimation of individual risks. Topics will include building models with large numbers of rating factors, interactions among rating factors, dealing with data issues like missing values, development on open claims, summarization and reality checks, and finally establishing the credibility of individual rating.
Source:
2006 Fall SIS- Predictive Modeling
Type:
concurrent
Panelists:
Daniel Finnegan, Catherine Eska
Practical Issues in Model Design
Textbook descriptions of model development are normally presented with rather straightforward examples. However, applying predictive modeling to insurance data presents a range of particular challenges. In this presentation some of the frequently encountered issues in developing a predictive model with insurance data and how to handle them will be discussed
Source:
2006 Fall SIS- Predictive Modeling
Type:
concurrent
Moderators:
Beniot Carrier
Panelists:
Charles Boucek, James Tanser
Combining GLM and Data Mining Techniques
Recent years have seen the introduction of a wide variety of data mining technologies into P&C insurance: neural nets, decision trees, support vector machines, and various other algorithms. In this session we will consider real-world examples of combining data mining results into sound rating systems using the GLM.
Source:
2006 Fall SIS- Predictive Modeling
Type:
concurrent
Panelists:
Greg Taylor, Daniel Finnegan
Visualizing predictive modeling results
The proper graphical presentation of a predictive model can be a critical diagnostic tool in the analysis of model results. Graphical presentation is also a key communication tool to individuals in an organization who do not have a detailed background in predictive modeling. The presenters will draw on their experience from a variety of predictive modeling projects in order to demonstrate a number of graphical presentation methodologies that they have found critical for properly diagnosing and presenting model results. Panelists will include techniques for understanding key aspects of the data, identifying and analyzing predictor variables, and summarizing key model results to senior management. Selected elements of the presentation will be in case study format.
Source:
2006 Fall SIS- Predictive Modeling
Type:
concurrent
Moderators:
Ann Griffith
Panelists:
Claudine Modlin
Data preparation
A crucial step in predictive modeling, data preparation is the most time consuming step in many projects. Data preparation involves exploring and cleaning data as well as augmenting data with externally and internally derived variables.
Source:
2006 Fall SIS- Predictive Modeling
Type:
concurrent
Moderators:
Ann Griffith
Panelists:
Louise Francis, Ravi Kumar
Special Challenges with large data mining projects
This session will present the issues that a project manager will likely face on a predictive modeling project. The focus will be on how to address these issues in order to bring a modeling project to a successful completion.
Source:
2006 Fall SIS- Predictive Modeling
Type:
concurrent
Moderators:
Ann Griffith
Panelists:
Beth Fitzgerald, Gary Ciardiello
Predictive Modeling for Smaller Companies - Commercial
In this session, we will discuss some of the issues and potential advantages faced by smaller carriers in performing and implementing predictive modeling for commercial lines. These issues include data needs, competitive analyses, implementation, distribution, and regulatory aspects.
Source:
2006 Fall SIS- Predictive Modeling
Type:
concurrent
Panelists:
Rick Smith, Katherine Barnes
Predictive Modeling for Smaller Companies - Personal
Many larger companies have embraced the idea of predictive modeling and have had the volume of data to produce credible results. Small- to medium-sized companies may wonder how predictive modeling can help them, especially when they do not have the data volume of their larger competitors. This session will discuss why predictive modeling has become more important for small- to medium-sized insurers, and how these insurers address some of the unique issues they face when developing models. The session will also cover some of the results obtained when applying these techniques and some of the unique advantages smaller companies have when approaching the predictive modeling process.
Source:
2006 Fall SIS- Predictive Modeling
Type:
concurrent
Panelists:
Roosevelt Mosley, Keith Toney
Claims/Agency Metrics
Claims-cost models that rely on the law of large numbers (most conventional actuarial models) become increasingly inadequate as the sample sizes of claims requiring estimation decrease. Correspondingly, the need for models that factor in the specific characteristics of those claims increases.
Source:
2006 Fall SIS- Predictive Modeling
Type:
concurrent
Panelists:
Shawna Ackerman, Greg Taylor
Modeling the settlement process for auto bodily injury liability claims
Auto, bodily injury, and liability claim settlements are negotiated between the insurer and the claimant or attorney. While economic damages such as medical bills and lost wages provide hard data upon which the settlement is based, the other major settlement component of general damages or "pain and suffering" often making up 60 percent or more of the settlement, is generally based on negotiated amounts that arise from a variety of hard and soft factors. This session will cover some recent research that establishes the key variables in a Tobit regression model of the settlement amount and provides a modeling framework for the negotiation process to reach that settlement. Parametric estimates of the settlement variables and negotiation process based upon Massachusetts auto data will be included.
Source:
2006 Fall SIS- Predictive Modeling
Type:
concurrent
Panelists:
Richard Derrig, Greg Rempala
Modeling Policyholder Retention
This session will cover the use of multivariate techniques to study and predict outcomes such as response rate, policyholder retention, and new business conversion. The panel will provide practical tips and illustrative results associated with modeling customer response data. Furthermore, the panel will address the benefits and applications of modeling customer response, in particular, the issues and opportunities associated with combining loss-cost models and customer-response models to determine optimal prices.
Source:
2006 Fall SIS- Predictive Modeling
Type:
concurrent
Panelists:
Michael Brockman, Duncan Anderson
Vehicle Symbol Development
The use of more refined vehicle rating plans in personal and commercial lines by several market innovators is beginning to cause companies to take a closer look at their own plans in order to make sure that their plans in conjunction with their more segmented rating plans are producing the most accurate rates. This session will discuss some of the plans being used by these insurers and discuss alternative vehicle rating systems which make better use of vehicle's individual characteristics.
Source:
2006 Fall SIS- Predictive Modeling
Type:
concurrent
Panelists:
Leroy Boison, Arthur Tabachneck
Predictive Modeling for Small Commercial Risks
In an attempt to replicate the successful applications in the personal lines industry, the commercial lines industry is speeding up its adoption of predictive modeling. This session will review the current predictive modeling development for small commercial risks. The session will also discuss the underwriting and pricing challenges for small commercial risks and how predictive modeling, such as scoring models, can address these needs. The efforts and challenges involved in building predictive models will be described, including data issues and analysis of models.
Source:
2006 Fall SIS- Predictive Modeling
Type:
concurrent
Panelists:
Beth Fitzgerald, Robert Walling
Homeowners
Many GLM ratemaking applications have focused on private passenger auto examples. This session will discuss how the nature of some homeowners' variables affects a predictive modeling analysis. These include both traditional rating variables (such as amount of insurance, deductible, and policy form) as well as external variables related to demographics or weather. The typical indivisible premium approach for analyzing homeowners' data does not lend itself well to proper investigation of these explanatory variables; therefore, the presentation will outline a case for modeling homeowners separately by peril.
Source:
2006 Fall SIS- Predictive Modeling
Type:
concurrent
Panelists:
Claudine Modlin