Search results

Sort By:

Limit search:

Bayesian Forecasting of the Discounted Payoff of Options on WIG20 Index in Discrete-Time SV Models

100%

Pajor A.

Acta Physica Polonica A

2008

vol. 114

issue 3

507-516

In this paper the bivariate stochastic volatility models (with stochastic volatility and stochastic interest rate) and the univariate fat-tailed and correlated stochastic volatility model (with stochastic volatility and constant interest rate) are used in the Bayesian forecasting of the payoff of European call options. The basic instrument is the WIG20 index. The predictive distribution of the discounted payoff is induced by the predictive distribution of the growth rate of the WIG20 index and the WIBOR1m interest rate. The Bayesian inference about the volatilities and the predictive distribution of the discounted payoff function is based on the joint posterior distribution of the latent variables, the parameters, and the predictive distribution of future observations, which we simulate via Markov chain Monte Carlo methods (the Metropolis-Hastings algorithm is used within the Gibbs sampler). The results show that allowing interest rate to be stochastic does not significantly improve forecasting performance of the discounted payoff. The predictive distributions of the discounted payoff are characterised by huge dispersion and thick tails, thus uncertainty about the future value of the payoff was ex-ante very big.

Bayesian Value-at-Risk and Expected Shortfall for a Large Portfolio (Multi- and Univariate Approaches)

63%

Pajor A., Osiewalski J.

Acta Physica Polonica A

2012

vol. 121

issue 2B

B-101-B-109

Bayesian assessments of value-at-risk and expected shortfall for a given portfolio of dimension n can be based either on the n-variate predictive distribution of future returns of individual assets, or on the univariate model for portfolio volatility. In both cases, the Bayesian VaR and ES fully take into account parameter uncertainty and non-linear relationship between ordinary and logarithmic returns. We use the n-variate type I MSF-SBEKK(1,1) volatility model proposed specially to cope with large n. We compare empirical results obtained using this (more demanding) multivariate approach and the much simpler univariate approach based on modelling volatility of the whole portfolio (of a given structure).

Refine search results

2 Acta Physica Polonica A

1 2012

1 2008

2 Pajor A.

1 Osiewalski J.

Search results

Bayesian Forecasting of the Discounted Payoff of Options on WIG20 Index in Discrete-Time SV Models

Bayesian Value-at-Risk and Expected Shortfall for a Large Portfolio (Multi- and Univariate Approaches)