Modelling stochastic volatility using hidden Markov models: a case study of the Kenyan securities Market
Date
2021
Authors
Bosire, Matilda Bosibori
Journal Title
Journal ISSN
Volume Title
Publisher
Strathmore University
Abstract
The biased parameter estimates generated by the Black-Scholes model have been attributed to the failure of the normality and constant volatility assumption to hold. Results of the improvement of the Black-Scholes-Merton model include time-varying volatility models which capture certain stylized facts of stock returns, with their use expected to improve the ability to price assets beyond the benchmarks provided by Black-Scholes. This thesis models stochastic volatility using Hidden Markov Models in Kenya. The univariate Stochastic volatility Model is calibrated to the Nairobi Securities Exchange 20 share index daily data for the period January 2012 to February 2021. The hidden Markov model (HMM) is employed in establishing volatility regimes while the Expected Maximization (EM) algorithm is employed in parameter estimation. Markov chain Monte Carlo (MCMC) and Sequential Monte Carlo (SMC) techniques are employed in filtering out noisy observations in parameter estimation. The 4-state model, which divides the economy into periods of very high, high, low, and very low volatility, is established to be optimal. Under each regime and filtering technique, different parameter estimates for single and multiple state models suggest a more dynamic framework for modeling the volatility process. The research findings contribute to theoretical literature on volatility-backed financial valuation and risk management in the context of regime switches.
Description
Research thesis submitted to Strathmore University in fulfillment of the requirements for the Master of Science in Mathematical Finance
Keywords
Hidden Markov models, Stochastic volatility, NSE20, Volatility regimes