1. The Impact of Predictive Analytics on Reserving
Brian Z. Brown, FCAS, MAAA
Principal and Consulting Actuary
Stan Smith
Predictive Analytics Consultant
April 29, 2016

2. Advancements In Reserving
Use of stochastic methods. Advancements in computing power have allowed for more sophisticated reserving methodologies
Distributions of loss reserves to better understand liabilities
Deterministic methods still dominant. Traditional development methods are still the go-to choice for most reserving actuaries
Methods are easier to explain to decision makers
Leaving something on the table? With the abundance of data being captured, could predictive analytics help us better understand our claims data in order to make better reserve projections?

3. Advantages of applying analytics in reserving
Strategic Reasons.
Competitive advantage
Operational Reasons.
Claims management/operations
Better information for pricing
Better information for reinsurance procurement
Financial Reasons.
Quicker closure rates (lower claims costs)
Lower defense costs
More accurate case reserves on higher value claims
Early intervention typically reduces claim costs

4. Predictive models for case reserves

6. Insurers cannot rest
“Companies that have not actively invested in improving their pricing sophistication, efficiency and risk management are at a competitive disadvantage and will not be relevant in the long term”.
Source: A.M. Best

7. Moneyball and Sabermetrics
Indicators of Offensive Success
Traditional Measure
Moneyball Measures
Home Runs
Batting Average
Stolen Bases
RBI’s
On-Base %-age
Slugging %-age
Pitch Data
Expected Future Runs Scored in an inning given certain conditions. ( data set)

8. Winning an unfair game
“People operate with beliefs and biases. To the extent you can eliminate both and replace them with data, you gain a clear advantage.”
Michael Lewis,
Moneyball: The Art of Winning an Unfair Game

9. Analytics can help identify “Useful” data
Leverage more of the data being captured
Traditional Approach
Big Data Approach
Analyzed information
All available information analyzed
All available information
Analyze small subsets of data
Analyze all data

10. Supervised versus Unsupervised approaches

11. SAMPLE KEY WORDS/PHRASES
Text mining variables
Text mining refers to the process of deriving relevant and usable text that can be parsed and codified into a word or numerical value.
Text mining can identify co-morbid conditions and/situations that will have profound impact on the outcome of a claim.
smoking
Pain unchanged
CXR
Diabetes/insulin/injections
Packs day/coughing
Pain killers/anti-depression
Children/school
Pain unchanged
Height/Weight
Homemaker wife went to work
c/o, CXR, FB, FX
CBT – Cognitive Behavior Therapy
SAMPLE KEY WORDS/PHRASES
Text sources: Adjuster notes, medical reports, independent medical exams, etc.

12. Modeling architecture
Data Store. All historical data collected and organized.
Training. Identifying company/internal/external data specific patterns.
Testing. Using “hold out” sets to measure the accuracy of predictions.

13. Virtual data warehouse
Improved Decisions
More Timely Decisions
Efficient Use of Resources
Improved Results
OPERATIONAL SYSTEMS
UNDERWRITNG
STRATEGIC
NEW Business
ETL
LOSS DATA
DATA
WAREHOUSE
Medical Vendor
Data
TACTICAL
Other
SCORING
ENGINE
BUSINESS
INTELLIGENCE
Milliman DB
REPORT
MONITOR
MEASURE
Closed Loop
Reporting
3RD PARTY
DATA ENRICHMENT

14. Complementary Analytics Solutions
Underwriting
Better Decisions
Leveraging data and analytics
Improved pricing and segmentation
Improved client targeting
WC Claim
Better Outcomes
Predictive Modeling - pro-active claim management
Data Warehouse - comprehensive view of internal data

15. Decision support example - claims
Quickly identify “creeping catastrophic” claims.
Less than 20% of claims cause 80% of losses
Create better claims outcomes with more timely and more detailed information.
Loss cost reductions that generally range from 3-6% per year
“Operationalize” into claims/medical protocols/rules.
Integrate management of all available sources of data/information.
“Second pair of eyes” on existing claim/medical vendors.
Ancillary benefits.
Data driven culture

16. Segmentation analysis
Divide All Claims into 5 buckets of 20% each.
After Scoring distribute by Risk Score
Highest Risk to the Right
Lowest Risk to the Left
Each Claim has a individual score
Worst Claim far right vs. Best Claim far left
Then add actual losses to test model accuracy
20%
20%
20%
20%
20%

17. Predictive modeling in action
Early ID < Day 30
Models Identify 20% of Claims that have 78% of total costs
High Risk
Medium Risk
Low Risk
11.47%
2.19%
4.74%
3.15%

19. Actuarial Dashboard – Auto Liability

20. Actuarial Dashboard – Incurred Loss & ALAE

21. Actuarial considerations
Changing development patterns. Nothing new for reserving actuaries, but important to monitor and be aware of implications.
Reserving
Pricing
Reinsurance
Communication with different business units is important

22. Initial measurements
Is the model identifying problematic claims earlier?
Impact of early intervention. Cost savings.
Impact on average case reserves.
Reserve strengthening
Adequate
Impact on payment pattern. Early intervention may result in high payments initially but lower total payments for a claim.

23. Actuarial considerations – after implementations
Adjustments to standard reserving methods
Berquist-Sherman
Varying / Adjusting assumptions
Loss Development Factors based on
Anticipated patterns to change
Measurement of how patterns are changing
Expected Loss Ratios based on
Projected cost savings

24. Predictive model case studies

25. Closed Claim Impact Client Case Study #1

26. Payment Trends – Totals over 10 Year Period

27. Payment Trends – Averages over 10 Year Period

28. Claim Age & Payment Trends – Model Impact

29. Reserving Impact Client Case Study #2

30. Reserving Trends – Model impact first 30 days

31. Reserving Trends all open claims – Model Impact

32. Actuarial considerations
Some models establish an ultimate per claim
Difference between model ultimate and incurred losses produces an indication of IBNR
How is pure IBNR determined?
Is the model ultimate an additional methods used for reserving?
How does the model ultimate change over time?
Documentation. Necessary so that another actuary may be able to review work product.
ASOP’s

33. Implementation considerations

34. Implementation planning is critical
Who has access to the model and its results? Access to the model and its output could potentially influence behavior
Could impact how case reserves are set in the future if claims adjusters have access to the model output
Impacts development patterns used for traditional actuarial reserving and ratemaking techniques
Would model still be relevant or applicable if underlying inputs are being changed?
Staff morale. Be cognoscente of how various business units might react to the implementation of the model
Intentions of model should be communicated to those who might be affected by the model
Fear of job elimination (claims adjusters, traditional reserving actuaries)
How to measure return on investment? Predictive modeling can be costly; make sure you’re getting the most out of your investment
Designate benchmarks before modeling
Regulation. Will regulators at some point need to review model?

35. Thank you