CDS Proxy Construction via Machine Learning Techniques Part II: Parametrization, Correlation, Benchmarking

Journal of Financial Data Science, Vol. No. 2, 2019

Posted: 6 Mar 2019 Last revised: 15 May 2019

See all articles by Raymond Brummelhuis

Raymond Brummelhuis

University of Reims Champagne-Ardenne

Zhongmin Luo

Birkbeck, University of London

Date Written: November 5, 2018

Abstract

This article is the second of two articles by the authors on the construction of CDS proxy rates. In the first article, the authors proposed a machine learning (ML) -based proxy-rate construction technique which uses classification to construct so-called Proxy-Names whose liquidly quoted CDS rates can be used as CDS proxy rates. The authors then compared the performances of ML classifiers from the eight most popular classifier families as function of carefully selected sets of feature variables. In this second article, the authors take a closer look at the performances of the individual classifiers as a function of their different parametrisations, which they refer to as an Intra-classifier performance study. The authors also examine the effects of feature variable correlations on classifier performance, and perform a benchmarking exercise by comparing the ML-based CDS-proxy technique with two currently used proxy-rate construction methods: Curve Mapping and Cross-Sectional Regression.

Keywords: CDS Proxy Construction, Counterparty Credit Risk, Machine Learning, Classification

JEL Classification: C4, C45, C63

Suggested Citation

Brummelhuis, Raymond and Luo, Zhongmin, CDS Proxy Construction via Machine Learning Techniques Part II: Parametrization, Correlation, Benchmarking (November 5, 2018). Journal of Financial Data Science, Vol. No. 2, 2019, Available at SSRN: https://ssrn.com/abstract=3334149

Raymond Brummelhuis

University of Reims Champagne-Ardenne ( email )

51096 Reims Cedex
France

Zhongmin Luo (Contact Author)

Birkbeck, University of London ( email )

Malet Street
London, London WC1
United Kingdom

Do you have negative results from your research you’d like to share?

Paper statistics

Abstract Views
493
PlumX Metrics