OPERATIONAL IT RISK FORECASTING AND ANALYSIS BASED ON BAYESIAN BELIEF NETWORKS

Abstract

This article provides the model for IT operational risk analysis, which is based on Bayesian networks. The model allows to predict IT risk losses depending on software quality, IT staff experience and utilized testing practices. The model is provided with hands-on example. In this example, predictive Bayesian inference and sensitivity analysis are performed to get a visual representation of the impact of different input variables on the IT operational losses. The abductive Bayesian inference is performed to analyze risk events and to localize root sources of these events. The model is implemented by means of RStudio and AgenaRisk tools. Results of the work can be used in practical work of banks and its technical departments to predict IT operational losses.

Keywords

• operational risk management
• it incident
• software testing
• operational value at risk
• bayesian network
• sensitivity analysis
• bayesian inference