Ruark Behavioral Analytics Advisory Council - next call June 23 at 2pm Eastern

Our next meeting is coming up soon.

This is by invitation only, so please RSVP if you have not done so already, or contact us if you are interested.  We plan to share 3 key things:

  • Starting the modeling process with relevant industry data, then tailoring to each company, typically leads to better assumption models than what an individual company can do using only its own data. We will demonstrate this with model factor selection, standard error terms for coefficient estimates, and actual-to-expected ratios.
  • Integration of models across behaviors is important. We will demonstrate this by showing the impact of partial withdrawal history on surrender behavior.
  • In collaboration with the UCONN Goldenson Center for Actuarial Research, we have developed a very powerful modeling approach for partial withdrawal behavior, particularly for GLWBs and the like, reflecting the impact of partial withdrawal history as well as behavioral variations between large ad hoc withdrawals and true income-related withdrawals. We will outline our process for developing these models and how to tailor them to each company, and share sample output.

Friendly reminder that our 2017 FIA industry studies and spring VA industry studies have been completed and are available for purchase, along with related assumption model development and benchmarking services.  If you are a data-contributing client and know what you would like to purchase, please contact us and we will arrange delivery and discussion. Or if you have any questions that would aid in your decision-making, please let us know. Our 2017 service and pricing summary can be found here.

Our 2017 plans for behavioral analytics



Are your assumptions informed by credible industry experience?


checkWe provide a powerful combination of industry- and company-level experience studies, predictive modeling, traditional analytical techniques, and expert judgment, based on seriatim monthly data since 2007, covering approximately 70% of the annuity industry.

Is your analytical framework robust as new data emerges?

checkWe work with you to customize and implement our behavioral analytics framework, with transparent linkage from experience data to assumption models, naturally suited to regular updates for inforce and new business.

Are your analytics granular enough to mitigate anti-selection and proactively manage changes in your business mix?

checkWe address the many factors of influence and their changes over time, including product and guarantee type, surrender charge period and duration, moneyness of guarantees on actuarial and nominal bases, contract size, tax status, age, gender, distribution channel and compensation structures, and income utilization and efficiency.

Is speed important to you?

checkIt is to us too. We provide the timely and immediately actionable results you need to efficiently manage your company’s behavioral risks.


We aim to be the platform and industry benchmark for principles-based insurance data analytics and risk management.


2017 VA and FIA Behavioral Modules

For each of: Options
VA Surrenders       VA Income Utilization       VA GMIB Annuitization       FIA Surrenders       FIA Income Utilization 1 2 3
Experience Studies – industry results in aggregate, along with your company results, in a detailed report with numerical exhibits covering key factors, cohorts, and dynamics
Customized assumption model – initially calibrated to industry results, and tailored to your company based on credibility techniques
Review of your current assumptions, and comparison to the customized model above
Benchmarking of your results relative to peers
Presentation and discussion with our team
Membership on our Behavioral Analytics Advisory Council




Would you like to learn more about implementation and pricing?


Timothy Paris

How well can you model GLWB behavior?

In collaboration with UCONN's Goldenson Center for Actuarial Research, we have developed a remarkable approach to modeling policyholder behavior for deferred annuities (particularly GLWB income utilization), with strong goodness of fit to historical data and high predictive power. All of this is fueled by industry-level data, and then tailored to each company using credibility procedures. Actuaries, A/E ratios in the 99-101% range?! Yes, it can be done. Hit us up.

Join us at the 2017 Retirement Industry Conference

Tim Paris will be moderating the session "Leadership Perspectives: Industry Benchmarking, Behavioral Analytics, and Executive Communication". Hope to see you there.

Podcast - Growing Your Company’s Predictive Analytics Capabilities.

Steve Wright and Tim Paris participated in this podcast of the Society of Actuaries' Predictive Analytics and Futurism section.

Predictive Analytics Case Study – Growing Your Company’s Capabilities

February 22, 2017

Length: 16:11

Sponsor: Predictive Analytics and Futurism Section

Description: Listen in as Shea Parkes, FSA, MAAA interviews Timothy Paris, FSA, MAAA and Steve Wright, FSA, MAAA from Ruark Consulting.  Tim and Steve have developed predictive analytics expertise at their company over the last few years.  They share stories from that journey and some suggested resources for others wanting to do the same.

Tim Paris sharing policyholder behavior insights in London


Non-traditional Life Insurance Products with Guarantees:From Variable Annuities to CPPI

8th - 9th March 2017, London

Tim Paris will be sharing policyholder behavior insights from the US market at the Non-traditional Life Insurance Products with Guarantees event next week in London.

National Retirement Sustainability Index

Updated for 2015 data. Taking a broader perspective beyond our usual focus on policyholder behavior analytics, here is a great reminder from UCONN's Goldenson Center for Actuarial Research about why annuities and other retirement savings and income products are so important.

New book: Non-traditional Life Insurance Products with Guarantees

Ruark CEO Timothy Paris, contributed chapter 9 (modeling and managing policyholder behavior risks) to this new book. Even better, all royalties are pledged to UNICEF. If your company is a client of Ruark, we are glad to subsidize your purchase -- please contact us for a 35% discount code.

Great stocking stuffer for yourself or beloved actuaries and quants!

SOA Annual Meeting presentation slides are available for download

soa2016-annual-mtgSOA Annual Meeting presentation slides are available for download

Thursday, November 3, 2016

Tim Paris presented and was a moderator at the 2016 SOA Annual Meeting.  Included are links to the engaging sessions Tim participated in.


Session 40 Panel Discussion: How Would I Get Started With Predictive Modeling?

Competency: Technical Skills & Analytical Problem Solving
Moderator(s): Douglas T. Norris, FSA, MAAA
Presenter(s): Timothy S. Paris, FSA, MAAA; Sandra Tsui Shan To, FSA, MAAA; Qinqing (Annie) Xue, FSA, CERA, MAAA
View presentation pdf.


Session 97: Variable Annuity Guaranteed Living Benefit Utilization Studies and Benefit Utilization in Fixed Indexed Annuities.

Competency: External Forces & Industry Knowledge
Moderator(s): Patrick David Nolan, FSA, MAAA
Presenter(s): Jafor Iqbal; Joseph E. Montminy, ASA, MAAA; Timothy S. Paris, FSA, MAAA
View presentation pdf.


Session 83: Panel Discussion: Product Development Impact of the Department of Labor Fiduciary Rule

Competency: External Forces & Industry Knowledge
Moderator(s): Timothy S. Paris, FSA, MAAA
Presenter(s): Colin Devine, CFA, CFP, CMA; Daniel J. Rueschhoff, FSA, MAAA; Yuan Tao, FSA, MAAA
View presentation pdf.