Likelihood-based inference for correlated diffusions

Kalogeropoulos, Konstantinos and Dellaportas, Petros and Roberts, Gareth O. (2007): Likelihood-based inference for correlated diffusions.

Abstract

We address the problem of likelihood based inference for correlated diffusion processes using Markov chain Monte Carlo (MCMC) techniques. Such a task presents two interesting problems. First, the construction of the MCMC scheme should ensure that the correlation coefficients are updated subject to the positive definite constraints of the diffusion matrix. Second, a diffusion may only be observed at a finite set of points and the marginal likelihood for the parameters based on these observations is generally not available. We overcome the first issue by using the Cholesky factorisation on the diffusion matrix. To deal with the likelihood unavailability, we generalise the data augmentation framework of Roberts and Stramer (2001 Biometrika 88(3):603-621) to d-dimensional correlated diffusions including multivariate stochastic volatility models. Our methodology is illustrated through simulation based experiments and with daily EUR /USD, GBP/USD rates together with their implied volatilities.

• Item Type: MPRA Paper
• Language: English
• Keywords: Markov chain Monte Carlo, Multivariate stochastic volatility, Multivariate CIR model, Cholesky Factorisation
• Subjects: C - Mathematical and Quantitative Methods > C1 - Econometric and Statistical Methods and Methodology: General > C13 - Estimation: General; G - Financial Economics > G1 - General Financial Markets > G12 - Asset Pricing ; Trading Volume ; Bond Interest Rates; C - Mathematical and Quantitative Methods > C1 - Econometric and Statistical Methods and Methodology: General > C15 - Statistical Simulation Methods: General; C - Mathematical and Quantitative Methods > C1 - Econometric and Statistical Methods and Methodology: General > C11 - Bayesian Analysis: General

References

\bibitem[A\"{i}t-Sahalia, 2005]{ait05} A\"{i}t-Sahalia, Y. (2005). \newblock Closed form likelihood expansions for multivariate diffusions. \newblock {\em Annals of Statistics}. \newblock To appear.

\bibitem[A\"{i}t-Sahalia and Kimmel, 2005]{ait:kim05} A\"{i}t-Sahalia, Y. and Kimmel, R. (2005). \newblock Maximum likelihood estimation for stochastic volatility models. \newblock {\em Journal of Financial Economics}. \newblock To appear.

\bibitem[Beskos et~al., 2006a]{bes:pap:rob:f06} Beskos, A., Papaspiliopoulos, O., Roberts, G., and Fearnhead, P. (2006a). \newblock Exact and computationally efficient likelihood-based estimation for discretely observed diffusion processes (with discussion). \newblock {\em Journal of the Royal Statistical Society: Series B (Statistical Methodology)}, 68(3):333--382.

\bibitem[Beskos et~al., 2006b]{bes:rob:stu:vos06} Beskos, A., Roberts, G.~O., Stuart, A., and Voss, J. (2006b). \newblock A {M}{C}{M}{C} method for diffusion bridges. \newblock Submitted.

\bibitem[Chernov and Ghysels, 2000]{che:ghy00} Chernov, M. and Ghysels, E. (2000). \newblock A study towards a uinified approach to the joint estimation of objective and risk neutral measures for the purposes of options valuation. \newblock {\em Journal od Financial Economics}, 56:407--458.

\bibitem[Chib et~al., 2005]{ch:pit:she05} Chib, S., Pitt, M.~K., and Shephard, N. (2005). \newblock Likelihood based inference for diffusion models. \newblock Submitted.

\bibitem[Cox et~al., 1985]{cir85} Cox, J.~C., Ingersoll, J.~E., and Ross, S.~A. (1985). \newblock A theory of the term structure of interest rates. \newblock {\em Econometrica}, 53:385--407.

\bibitem[Daniels and Kass, 1999]{dan:kas99} Daniels, M. and Kass, R. (1999). \newblock Nonconjugate bayesian estimation of covariance matrices in hierarchical models. \newblock {\em Journal of the American Statistical Association}, 94:1254--1263.

\bibitem[Delyon and Hu, 2007]{del:hu07} Delyon, B. and Hu, Y. (2007). \newblock Simulation of conditioned diffusions and applications to parameter estimation. \newblock {\em Stochastic Processes and Application}. \newblock To appear.

\bibitem[Durham and Gallant, 2002]{dur:gal02} Durham, G.~B. and Gallant, A.~R. (2002). \newblock Numerical techniques for maximum likelihood estimation of continuous-time diffusion processes. \newblock {\em Journal of Business \& Economic Statistics}, 20(3):297--316. \newblock With comments and a reply by the authors.

\bibitem[Elerian et~al., 2001]{ele:ch:she01} Elerian, O.~S., Chib, S., and Shephard, N. (2001). \newblock Likelihood inference for discretely observed non-linear diffusions. \newblock {\em Econometrica}, 69:959--993.

\bibitem[Eraker, 2001]{era01} Eraker, B. (2001). \newblock Markov chain {M}onte {C}arlo analysis of diffusion models with application to finance. \newblock {\em Journal of Business \& Economic Statistics}, 19(2):177--191.

\bibitem[Ghysels et~al., 1996]{ghy:har:ren96} Ghysels, E., Harvey, A., and Renault, E. (1996). \newblock Stochastic volatily, in. \newblock {\em Handbook of Statistics 14, Statistical Methods in Finance}. \newblock G.S. Maddala and C.R. Rao (eds), North Holland, Amsterdam.

\bibitem[Golightly and Wilkinson, 2006]{gol:wil06} Golightly, A. and Wilkinson, D. (2006). \newblock Bayesian sequential inference for nonlinear multivariate diffusions. \newblock {\em Statistics and Computing}, 16:323--338.

\bibitem[Golightly and Wilkinson, 2007]{gol:wil07} Golightly, A. and Wilkinson, D. (2007). \newblock Bayesian inference for nonlinear multivariate diffusions observed with error. \newblock {\em Computational Statistics and Data Analysis}. \newblock In press.

\bibitem[Heston, 1993]{hes93} Heston, S. (1993). \newblock A closed-form solution for options with stochastic volatility. with applications to bonds and currency options. \newblock {\em Review of Financial Studies}, 6:327--343.

\bibitem[Hull and White, 1987]{hul:whi87} Hull, J.~C. and White, A.~D. (1987). \newblock The pricing of options on assets with stochastic volatilities. \newblock {\em Journal of Finance}, 42(2):281--300.

\bibitem[Jones, 1999]{jon99} Jones, C.~S. (1999). \newblock Bayesian estimation of continuous-time finance models. \newblock Unpublished paper, Simon School of Business, University of Rochester.

\bibitem[Kalogeropoulos, 2007]{kal07} Kalogeropoulos, K. (2007). \newblock Likelihood based inference for a class of multidimensional diffusions with unobserved paths. \newblock {\em Journal of Statistical Planning and Inference}, 137:3092--3102.

\bibitem[Kalogeropoulos et~al., 2007]{kal:rob:del07} Kalogeropoulos, K., Roberts, G., and Dellaportas, P. (2007). \newblock Inference for stochastic volatility models using time change transformations. \newblock Submitted.

\bibitem[Pedersen, 1995]{ped95} Pedersen, A.~R. (1995). \newblock A new approach to maximum likelihood estimation for stochastic differential equations based on discrete observations. \newblock {\em Scandinavian Journal of Statistics. Theory and Applications}, 22(1):55--71.

\bibitem[Pinheiro and Bates, 1996]{pin:bat96} Pinheiro, J. and Bates, D. (1996). \newblock Unconstrained parametrizations for variance-covariance matrices. \newblock {\em Statistics and Computing}, 6(3):289--296.

\bibitem[Roberts and Stramer, 2001]{rob:str01} Roberts, G. and Stramer, O. (2001). \newblock On inference for partial observed nonlinear diffusion models using the metropolis-hastings algorithm. \newblock {\em Biometrika}, 88(3):603--621.

\bibitem[Rogers and Williams, 1994]{rog:wil94} Rogers, L. C.~G. and Williams, D. (1994). \newblock {\em Diffusions, Markov processes and martingales, 2, Ito calculus}. \newblock Wiley, Chicester.

\bibitem[S{\o}rensen, 2004]{sor04} S{\o}rensen, H. (2004). \newblock Parametric inference for diffusion processes observed at discrete points in time: a survey. \newblock {\em International Statistical Review}, 72(3):337--354.

\bibitem[Stein and Stein, 1991]{ste:ste91} Stein, E.~M. and Stein, J.~C. (1991). \newblock Stock proce distributions with stochastic volatility: an analytic approach. \newblock {\em Review of Financial Studies}, 4(4):727--752.

\bibitem[Tanner and Wong, 1987]{tan:won87} Tanner, M.~A. and Wong, W.~H. (1987). \newblock The calculation of posterior distributions by data augmentation. \newblock {\em Journal of the American Statistical Association}, 82(398):528--540.