Risk Analytics
Sharpen risk measurement capabilities to improve the portfolio performance
Overview
We combine proprietary tools, methodologies and analytical techniques with expertise in risk management to help clients optimize their risk exposures, improve risk adjusted performance, reduce losses, and optimize capital allocation across businesses.
Risk analytics is evolving rapidly as the sources of data and complexity of data are increasing exponentially. Organizations are investing to harness their massive and rapidly growing quantities of information and new types of data coming from a wide variety of external sources. Given the availability of unprecedented streams of data, it is but natural to invest in building better models with greater predictive power by using advanced techniques.
We help our clients to develop and maintain the cutting-edge risk analytics framework. Our risk analytics offerings span across the risk types.
Approach
- Understand the Existing Portfolio: Understand the existing portfolio of the financial institution. Identify the distinct risk segments for model/scorecard development.
- Identify & Prepare Data: Identify the data required for modelling, both internal as well as external data. Collect, clean and prepare the data for modelling. We ensure that data sample is representative of the population and any bias in data is well articulated and quantified.
- Identify Techniques: Based on type of data & its dimensions such as volume, variety and velocity of data.
- Develop Model: Establish factor long list, conduct statistical analysis to short list the factors, finalize the factors based on their power and correlation with other factors, develop model based on the finalized factors, compare alternative models, use domain knowledge to short list the final model.
- Implement: Implement the model in secure environment for ongoing use in decision making and to collect data for evaluation of the model performance.
- Model Maintenance: Evaluate its performance on an ongoing basis, upgrade/modify the model if its performance is dipping, incorporate new sources of data or new factors to improve the performance.
- Documentation: Comprehensive documentation of the model development methodology. We ensure that all statistics, methodology and decisions are clearly articulated in the documentation for any 3rd party review.
