Smarter maths generates better value insurance products

Supplementary content information

Research into data mining and statistical analysis by researchers at the University of East Anglia (UEA), has created more competitive insurance products while pioneering key advances in computer science and statistics for industry.

  • Research led to key advances in computer science and statistics for industry
  • Helped generate more competitive quotes for Aviva’s general insurance and pensions customers
  • Enabled savings of many millions as a result of changes in pricing for general insurance products
  • Saved £150,000 saved through an in-house risk costing project

In collaboration with UK insurance group, Aviva (formerly Norwich Union), the research has led to considerably improved Aviva products and more competitive quotations for customers while maintaining the company’s profitability. It also helped pioneer data mining techniques today relied on by a host of industries and sectors, from banking to retail, security to social sciences.

The research is now led by Professor Elena Kulinskaya, Aviva Chair in Statistics at UEA, who has used it to tackle specific challenges relating to medical statistics and the insurance industry, such as customers’ future longevity.

Professor Kulinskaya’s work builds on EPSRC and Norwich Union-supported studies in the 1990s led by Professor Vic Rayward-Smith, a pioneer in the emerging discipline of data mining. Using advanced computer algorithms, data mining makes it possible to identify potentially useful patterns, such as customer preferences, from vast and complex data sets.

Professor Rayward-Smith and Dr Beatriz de la Iglesia, then a Postdoctoral Research Associate, were among the first to describe how data mining could be used in industrial applications. With the advent of the Big Data era, data mining and tools such as those developed by the UEA team are now essential to identify patterns and future trends across a wide range of sectors, from the insurance industry to healthcare; telecommunications to retail.

The managing director of Aviva Life’s At Retirement division acknowledges the value of the UEA research. He says: “Insurance is a very competitive industry and correct pricing and marketing are core to our survival. It’s difficult to quantify the exact savings that UEA’s input has made to Aviva, but it would certainly run into many millions”

Find out more

After completing her doctorate, in the 1990s, Dr Beatriz de la Iglesia joined the academic staff in UEA’s School of Computing Sciences, specialising in data mining algorithm development and application of data mining techniques for the financial, biological and medical sectors. In 2004, a three-year EPSRC first grant enabled her to build on this research through the creation of new ‘meta-heuristic’ computer algorithms.

Professor Elena Kulinskaya has taken this research further. Working with Aviva staff, the UEA School of Computing Sciences team has focused on two main areas of Aviva’s business – General Insurance and Annuities.

General Insurance:

Sophisticated ‘scoring’ algorithms developed by Professor Rayward-Smith and Dr Beatriz de la Iglesia have enabled Aviva to increase the competitiveness of its quotations for both pricing and marketing in general insurance, such as car insurance. Adoption of these techniques has allowed a more competitively-priced product to be offered to customers while maintaining Aviva’s profitability.

Aviva is also using data mining techniques devised by the UEA team to develop and implement predictive models for targeted marketing campaigns, helping the company to improve its capability and effectiveness.

The use of these techniques greatly enhanced the training of Aviva data analysts via an in-house MSc programme developed by the UEA team specifically for the company. Over 100 Aviva staff have taken part in the scheme.

Annuities:

Professor Kulinskaya led a joint UEA and Aviva team investigating the main insurance risks in the three most prevalent chronic medical conditions within the annuities market. The research enabled Aviva to generate quotations in house and avoid an estimated £150,000 in consultancy costs associated with outsourced customer quotations determined without Aviva control or contribution. It also helped the company ensure its guarantees to customers are sound, robust and correctly priced.

Professor Kulinskaya has continued to build on this early research with funding from the Institute and Faculty of Actuaries to develop novel statistical and actuarial methods for modelling mortality and trends in morbidity. The multidisciplinary team of actuaries, statisticians, health economists, computer scientists and medical researchers will be establishing the drivers for the increasing longevity seen in developed and developing countries in order to predict how they may change over time and how this would affect life expectancy. (add link: http://www.bighealthactuarialdata.ac.uk/home