Validus Re Uses Revolution R Enterprise for Risk Management
Validus Re is a global company whose underwriting philosophy is to maximize its return on equity subject to prudent risk constraints on the amount of capital its exposes to loss events. This underwriting philosophy runs alongside its adherence to its enterprise risk management strategy and the maintenance of a conservative investment approach which focuses on preserving capital and liquidity. Validus Re underwrites a balanced and well-spread portfolio of primarily short-tail reinsurance business giving superior risk and return characteristics. In this way, it aims to create franchise value while maximizing sustainable long-term growth in shareholder value.
Reinsurance is a highly complex business due to the detailed data analysis required in order to underwrite a broad variety of risks subject to a broad array of risk perils including weather and other elements, fire, financial volatility or man-made action such as terrorism. These risk perils can vary in both severity and frequency.
Actuarial teams within a reinsurance company have a variety of responsibilities that have a direct impact on the company’s financial health and quality rating by third party agencies. They must work with the underwriting department to estimate risk and establish prices for individual opportunities. They also measure and manage the overall risk being carried by the company, which is called economic capital modeling, which takes into account risk from the company’s investment portfolio, operational risk, credit risk and its reinsurance risk. Economic capital modeling must take into account rating agency capital requirements (e.g. AM Best, Regulators’ capital requirements, etc.) whereby the company must meet established standards related to its solvency.
In order to manage risk actuaries must analyze and correlate worldwide risk factors and measure their potential impact on the company’s portfolio. According to Jamie Botelho, Vice President and Pricing Actuary, everything in insurance is magnified at the reinsurance level and cyclical losses are common. He explains, "The biggest overall challenge is that the answers we seek are abstract. There are accepted methods, but they all have the challenge of uncertainty."
"What tests the company’s model is how well it handles the correlation between the risk factors so we can efficiently spread risk worldwide. The fact is when hurricane hits, it affects many risks, which is why any company would limit its exposure in a hurricane zone such as South Florida. Yet not all events or causes of loss are correlated through geography alone and some have far-reaching effects. R helps me correlate the risk factors that are not simply event-related."
He continues, "Validus Re is a young company, without decades of data that would provide an easier look into how things move together within our portfolio. This makes the strength of the work we do with Revolution R Enterprise even more important."
Like many companies, Validus Re’s environment for advanced analytics is complicated by a large amount of data coming from many sources as well as the need to use as model inputs the outputs from third party applications, such as underwriting financial systems
There are different models in use for various risk-related processes and departments throughout the company. Botelho uses Revolution R Enterprise, which is based on the open source statistical programming language R, to bring together output from these as well as from third parties such as RMS to model risk for the entire company. From an underwriting perspective, he creates a tremendous amount of data when running simulations based on 100,000 years of events, which he does so that he can see for example what would be affected if there were to be a hurricane in year 592. The amount of data would be too much for Excel, yet Revolution R Enterprise handles it easily, he says.
Botelho explains, "There are add-ins for Excel that allow you to model copulas and generate random numbers, yet they are limited. What I like about Revolution R is that it is based on a programming language so I can do things in a certain order. I can make my own functions that do exactly what I want to do. If I have more than a few million rows of data, it becomes unwieldy in Excel. With Excel, I can do 100,000 simulations 20 minutes, whereas with Revolution R it can be done in seconds."
Botelho continues, "This approach is more advanced than basic correlations because it allows me to define where the correlation happens. As things become more and more extreme, they become more and more correlated and I need a model that can estimate my risk in a way to that enhances our confidence in our pricing and reserving. Modeling with Revolution R Enterprise gives me that."
Even for more basic analysis, Botelho can work efficiently in Revolution R Enterprise. " What takes a day to do in Excel can be done in 15 minutes by writing a few lines of R code in Revolution R Enterprise."
Botelho says that he is now able to make decisions based on models that take into consideration all of the factors and data that he wants to include. "I don’t have the limits I once had. We know that tail events are more correlated than events in the middle of the structure. We are getting more accuracy. Therefore, we are better reflecting the risk in our judgment."
The company is so confident in the models generated in Revolution R Enterprise that it is used to test the results that are generated by the company’s vendor production system. "It is a closed system and we don’t really have visibility into it. If I model the output in Revolution R and it comes out different, which has indeed happened, I work with the actuarial team to look for inconsistencies between what R tells us and the system logic in the vendor production system and reconcile the differences. It has helped identify errors in logic as well as updating and inputting data. Revolution Analytics is great for ad hoc analysis – to mock up what we want to implement in the commercial applications, which roll up to the holding company’s systems," he explains.
The executive team is very happy with the results and understands the complexity of the economic capital modeling project. "We are working in a new, more accurate and inclusive way, which enhances our confidence in our judgments," he says.
About Validus Re
Validus Reinsurance, Ltd. was founded in 2005 and is an operating subsidiary of Validus Holdings, Ltd. Headquartered in Bermuda, Validus Re is a global provider of short-tail lines of reinsurance including property catastrophe, property pro rata and property per risk, as well as marine and energy, and other specialty lines.
About Revolution Analytics
Revolution Analytics is the leading commercial provider of software and services based on the open source R project for statistical computing. The company brings high performance, productivity and enterprise readiness to R, the most powerful statistics language in the world. The company’s flagship Revolution R Enterprise product is designed to meet the production needs of large organizations in industries such as finance, life sciences, retail, manufacturing and media. Used by over two million analysts in academia and at cutting-edge companies such as Google, Bank of America and Acxiom, R has emerged as the standard of innovation in statistical analysis. Revolution Analytics is committed to fostering the continued growth of the R community through sponsorship of the Inside-R.org community site, funding worldwide R user groups and offering free licenses of Revolution R Enterprise to everyone in academia.