Operational risk is one of the key focus areas of Basel II, Solvency II, Sarbanes-Oxley and other regulatory guidelines. We help institutions in building operational risk measurement framework with our expertise in operational risk modelling. We help institutions design and implement loss data collection framework and scenario analysis framework. We provide extensive support in implementation, vendor selection and UAT of AMA modelling software. We pay particular attention to parameter uncertainty involved in fitting probability distributions to loss data that may lead to unreliable capital figures, data adequacy, model complexity, biases in parameter estimation, stability of VaR numbers post convolution of frequency and severity distributions, aggregation modelling, checking applicability of EVT to data to avoid over-estimation of capital requirement and capital stress testing.
We help institutions in building, implementing and maintaining framework and technical platform for collection of loss data. Detailed data is collected about operational risk events, loss, recoveries, costs, controls, linkages with other events etc.
Scenario analysis framework using percentile approach, interval approach and individual scenario approach is covered in our SBA framework.
We help institutions to quantify uncertainty involved in parameter estimation during fitting of probability distributions to loss data. This is achieved through parametric as well as non-parametric bootstrapping as well as through analysing variance of MLE estimates. Adequacy of loss data is a key factor that drives parameter uncertainty and key input in deciding the weights assigned to loss data and scenarios.
We assist institutions in quantifying model complexity, which is then used for selecting best fitting distribution to loss data.
We assist in performing OpVaR computations using multiple methods such as Monte Carlo Simulation, Closed-form solution and Fast Fourier Transform, with and without incorporating insurance mitigation. To understand reasonableness of capital estimates, empirical simulations are also performed. Aggregation Modelling is performed using correlation matrix and user selected copula that decides tail dependence (t copula and gaussian copula).
We assist in examining applicability of EVT to loss data as otherwise it may lead to over-estimation of capital requirement. For cases where EVT is applicable, we help in selecting an appropriate threshold for fitting EVT distributions to draw a balance between parameter estimation bias and uncertainty.
We have conducted Operational Risk Modelling Analytics workshops (Excel and R software based) for operational risk analysts and modelling teams.
We perform UAT and independent validation of OpVaR Modelling framework and Software for banks to ensure operational risk measurement framework is in line with Basel II AMA requirements.