Pricing Convertible Bond Based on Integration of Support Vector Machine and Copula Function

As a kind of hybrid financial instrument, the pricing of convertible bond (CB) has constituted a great challenge. This paper developed a novel method integrating support vector machine (SVM) with copula function. Different from existing single-factor or bifactor pricing models based on corporate value and the underlying stock price respectively, this model can effectively deal with many constrains on the CB pricing, such as nonlinearity, departure from normality, multivariate joint distribution, variable dependence structure, and so on. In particularly, the new model exhibited great flexibility in that copula function can portray dependence structure between the underlying stock price and interest rate, and that SVM can further tackle nonlinear relationship among variables. Empirical analysis showed that the proposed model enhanced generation ability of out-of-sample, with mark increase in CB pricing accuracy compared with the single SVM, and that the CB value sensitivity to the underlying stock and the dependence structure is also measured handily and effectively through the model.