According to the FRSGlobal, the in-built backtesting techniques of the solution will allow risk managers to benefit from the application of statistical analysis on profit & loss data compared against VaR measures at configurable confidence levels, holding periods and sample sizes.
The advanced VaR backtesting techniques will help risk managers to better validate model accuracy statistically and to trace exceptions with classified quantitative information along the criteria outlined in Basel Committee framework on backtesting.
The incorporation of binomial test statistics into the FRSGlobal’s enhanced VaR backtesting techniques, will help risk managers return the probability of observing the encountered exception at a given confidence level and number of trials.
Kupiec test statistics are also built in, which will be used to help determine whether the observed frequency of exceptions is consistent with the frequency of expected exceptions according to the VaR model and chosen confidence interval.
The RiskPro solution of the FRSGlobal covers a wide scope and depth of financial analysis, helping ensure consistency of results and reduction in the costs of analysis.
The solution includes a flexible modular software platform that helps to satisfy the internal and external analysis requirements in financial risk analysis – for small to large organizations; financial product coverage from saving accounts, complex loans and insurance instruments to exotic options and structured products.
It also offers implementation and support resources, providing quality services, knowledge-transfer and effective solution implementation, use and support and global accessibility and delivery capability through local presence and international teams.
FRSGlobal head of Product Management Thomas Brouwer said that as risk managers become more accountable for the risk models used, they will no longer be able to say that their VaR models incorporate too many simplifications and assumptions in order to measure the risk appropriately.
"More and more they will be urged to strive for VaR models that capture risks adequately and backtest models more stringently," Brouwer said.
