An app on the new Bloomberg App Portal, Savvysoft OTC Backtesting&Risk streamlines the backtesting and risk assessment of illiquid OTC derivatives and To-Be-Announced OTC derivatives structures. For the first time ever, Savvysoft’s top-ranked derivatives models have been seamlessly integrated with Bloomberg’s wealth of historical derivatives data resulting in backtesting and risk assessment that are orders of magnitude faster than what currently exists.With OTC Backtesting&Risk, you can generate historical time series of mark-to-model OTC derivatives prices almost instantaneously.
OTC Backtesting&Risk is a boon for risk managers, derivatives structurers, and derivatives investors, saving them countless hours of time spent setting up their historical analysis and generating historical prices. Use OTC Backtesting&Risk to rapidly calculate not only VaR (Value at Risk), and CVA (Counterparty Credit Adjustments), but also to perform the what-if analysis that is so critical in zeroing in on optimal structures for the trading of OTC instruments.
OTC Backtesting&Risk allows OTC derivatives users to instantly view the historical performance of a wide variety of plain vanilla and complex, illiquid derivatives in the Equity, Interest Rate, FX and Commodity markets, including Historical VaR and CVA. Because one of the product’s core features is specifically designed to generate a historical time series of OTC derivatives prices, users are able to save hours of time setting up their analysis, performing the task in seconds, not hours.
OTC Backtesting&Risk automatically taps into Bloomberg’s vast database of price, yield, and volatility data, enabling users to call into Savvysoft’s award-winning TOPS pricing models by simply specifying a ticker for price and volatility or a currency code for rates and rate volatilities, making it faster than ever for market participants to quickly generate prices and sensitivities/Greeks for all types of derivatives. Users can even discount bond cashflows with credit curves such as US AA Bank with the appropriate code, providing an extension to Bloomberg’s existing set of discount curves.
Investors, salespeople, and structurers can quickly track the historical performance of any plain vanilla or complex derivative or structured note to assess the likely future behavior of the instrument
OTC Backtesting&Risk uses Savvysoft’s award-winning TOPS derivatives pricing models with Bloomberg data to generate a time series of mark-to-model prices for a wide array of illiquid OTC derivatives structures. Because these prices are generated at runtime, instead of being limited to exchange-traded closing prices, any instrument can be valued. OTC Backtesting&Risk can even derive a price history for a new issue which didn’t actually exist in the past. This time series of prices can then be analyzed, with detailed risk and return statistics automatically and expeditiously calculated, along with VaR and CVA.
The models inside OTC Backtesting&Risk are the same-award winning TOPS models which have been used by major market participants to price their books for nearly twenty years, so the price histories generated by the app are equivalent to where the instrument would have actually closed each day.
Backtesting and VaR can be generated for instruments that roll down over time, or have constant risk
Risk measurement of options and derivatives is complicated by the fact that these instruments have a clearly defined lifespan. Sometimes users want to look at how the instrument will perform given that its expiration is going to be getting shorter and shorter in the future, and sometimes they want to look at a time series as if the expiration is held constant over time.
Similarly, as the underlying price moves, the moneyness of the option changes (what is now in the money becomes out of the money as the underlying moves against the option holder).
OTC Backtesting&Risk offers users several choices for how to adjust expiration, maturity and strike over the historical backtesting period, allowing them full flexibility in defining the type of VaR they want to calculate.
Risk managers and traders can view VaR for individual instruments, as well as portfolios, trading strategies and customized structured products that are combinations of core component derivatives
The generation of historical performance allows users to calculate several statistics based on the results, including the mean return, the standard deviation of returns, and the percentiles of the frequency distribution of returns. The 5th-percentile worst case loss is the Value at Risk, or VaR, and this important risk measure is automatically calculated by the app.
The VaR that is calculated is the Historical VaR, which has several advantages over the other two major types, Parametric VaR and Monte Carlo VaR, most importantly no assumptions are made regarding the correlation of each asset with all the other assets.
The app can calculate VaR for a single instrument, as well as for a portfolio of instruments. Multiple portfolios can be stored, allowing VaR be measured on:
Credit risk managers and swaps clearing firms can calculate CVA based purely on historical data
OTC Backtesting&Risk employs a multi-period Monte Carlo simulation to generate future scenarios of instrument values over many time periods into the future. For example, a swap can be priced every quarter for the next 20 years, using one path of randomly generated interest rates. This can be repeated for many paths of quarterly swap values, which are then used to calculate CVA. The data driving the simulation is historical data, and all the advantages of Historical VaR over Monte Carlo, including not being required to estimate correlations, applies to OTC Backtesting&Risk’s CVA calculation.
The app also includes a calculator and both a 2-D graph and a 3-D interactive heat map.
To access OTC Backtesting&Risk on the Bloomberg App Portal
Bloomberg Professional service subscribers can instantly access the new Savvysoft app at
OTC Backtesting&Risk can be licensed for $499/month. A free trial of the app is available on Bloomberg.