Comparing Prediction Power of Artificial Neural Networks Compound Models in Predicting Credit Default Swap Prices through Black–Scholes–Merton Model

Abstract

Default risk is one of the most important types of risks, and credit default swap (CDS) is one of the most effective financial instruments to cover such risks. The lack of these instruments may reduce investment attraction, particularly for international investors, and impose potential losses on the economy of the countries lacking such financial instruments, among them, Iran. After the 2007 financial crisis, the importance of CDS has increasingly augmented because theoretically and practically, this instrument could significantly prevent catastrophes such as the mentioned crisis. The present study seeks to predict the price of CDS contracts with the Merton model as well as the compound neural network models including ANFIS, NNARX, AdaBoost, and SVM regression, and compare the predictive power of these algorithms which are among the most prestigious, intelligent models in finance. The research statistical population includes the A-rated North American and European companies which are known as the reference entities for credit default swaps. Data were collected from the Bloomberg Terminal for an eight-year period from 2008 to 2015. Contracts of 125 companies were selected as the statistical sample. The results reveal that the average predictive power of the NNARX is higher than that of other algorithms under scrutiny.