University of Twente Student Theses


Extended analysis of back testing framework for value at risk

Roekel, G.J. van (2008) Extended analysis of back testing framework for value at risk.

[img] PDF
Abstract:model and the back testing procedure are important parts of the banks market risk framework. The Value-at-Risk model provides a daily measure th exceed once every 100 days. DNB and Rabobank agreed to perform a periodic analysis of the VaR model that goes beyond the regulatory guidelines. Every quarter Rabobank International tests the accuracy of its VaR model using the regulatory back test. This back test checks the number of times the VaR was breached (called exception). Based on this number of exceptions this test judges if the VaR model is accurate or not. The regulatory back test has its limitations. Therefore, we conducted a literature research to investigate alternative back test methods. This resulted in a framework of five back tests that together test the most important properties of a VaR model: - exception frequency: the number of realised exceptions - exception clustering: independency of exceptions over the tested period. - exception size: the size of the exception We implemented the five back tests in a test framework that Rabobank International can use for the periodic back testing beyond regulation. <<deleted, confidential>>
Item Type:Essay (Master)
Rabobank International
Faculty:BMS: Behavioural, Management and Social Sciences
Subject:85 business administration, organizational science
Programme:Industrial Engineering and Management MSc (60029)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page