Contingent Convertible "CoCo" Bonds in Bank Risk Management Empirical Pricing ComparisonTools Kowlessur, Dhanishta (2016) Contingent Convertible "CoCo" Bonds in Bank Risk Management Empirical Pricing Comparison. [Dissertation (University of Nottingham only)]
AbstractThis work compares the performance of three pricing approaches on the CoCos issued on 21st of April 2014 by the China Merchants Bank (ISIN: CND100007RX8) through model fitting analysis with CoCo market prices. The bond specifications have been drawn from Moody’s CoCo Monitor Database. With an observation period from 10th of May 2014 to 31st of July 2015; a total of 284 trading days, the daily computed CoCo prices from the Credit Derivatives Approach, the Equity Derivatives Approach and the J.P. Morgan Model are determined using Moody’s Analytics. Based on RMSE and MASE measures, the J.P. Morgan Model is found to be significantly superior to the other valuation mechanisms while the Equity Derivatives Model is observed to be least-performing. Overall, the impact of integrating risk dynamics in the models, though positive except for the Equity Derivatives Model, is less pronounced probably because of a relatively sound financial environment prevailing,. The results of the Equity Derivatives Approach favoring the model in the absence of jumps were interpreted as either the ambiguous effect of option values or the addition of excessive risks in the valuation mechanism by the jump version model to the non-jump model. The assumption of constant conversion intensity for the Credit Derivatives Approach appears to be too simplistic and far-fetched from reality; empirical analysis supported a piecewise constant parameter instead. While the models fairly explain market prices, they generate lower CoCo prices than the actual market values. Three contingencies are subsequently inferred; the models allowing for more risks than actually required, the China CoCo market being culturally more risk-seeking, or the presence of an excessive demand for CoCos from yield-hungry investors. Further studies are anticipated for any possibility of generalization.
Actions (Archive Staff Only)
|
Tools
Tools
![[img]](/style/images/fileicons/application_pdf.png)
