Recent Developments <strong>in</strong> SeasonalVolatility Models 13Recent Developments <strong>in</strong> Seasonal Volatility Models43g ′′ [E(σt 2)] = √ 1 [ ⎛ ⎞⎝ E(σ2 t )+2(log(S/K)+rT)√⎠2π 4E(σt 2) E(σt 2)⎛⎞+ ⎝ 6(log(S/K)+rT)+E(σ2 √ t ) ⎠8[E(σt 2)]2 E(σt 2)( )[E(σ 2t )] 2 − 4(log(S/K)+rT) 2[E(σt 2)]2 ] {exp − (2(log(S/K)+rT) − }E(σ2 t ))28E(σt 2) ,where N denotes the st<strong>and</strong>ard normal CDF, <strong>and</strong> under the option pric<strong>in</strong>g model with seasonalGARCH volatility,E(σ 2 t )=κ (y) =ω() ,pP1 − ∑ φ i)(1 − ∑ Φ ii=1i=13∞ .3 − 2 ∑ Ψ 2 jj=17. Conclud<strong>in</strong>g remarksIn this chapter we propose various classes <strong>of</strong> seasonal volatility models. We consider timeseries processes such as AR <strong>and</strong> RCA with multiplicative seasonal GARCH errors <strong>and</strong> SVerrors. <strong>The</strong> multiplicative seasonal volatility models are suitable for time series whereautocorrelation exists at seasonal <strong>and</strong> at adjacent non-seasonal lags. <strong>The</strong> models <strong>in</strong>troducedhere extend <strong>and</strong> complement the exist<strong>in</strong>g volatility models <strong>in</strong> the literature to seasonalvolatility models by <strong>in</strong>troduc<strong>in</strong>g more general structures.It is well-known that f<strong>in</strong>ancial time series exhibit excess kurtosis. In this chapter we derivethe kurtosis for different seasonal volatility models <strong>in</strong> terms <strong>of</strong> model parameters. We alsoderive the closed-from expression for the variance <strong>of</strong> the l-steps ahead forecast error <strong>of</strong> i)y n+l <strong>in</strong> terms <strong>of</strong> ψ-weights <strong>and</strong> model parameters, <strong>and</strong> <strong>of</strong> ii) squared series Y n+l <strong>in</strong> terms<strong>of</strong> Ψ-weights, model parameters <strong>and</strong> the kurtosis <strong>of</strong> ɛ t . <strong>The</strong> results are a generalization <strong>of</strong>exist<strong>in</strong>g results for non-seasonal volatility processes. We provide examples for all the differentclasses <strong>of</strong> models considered <strong>and</strong> discussed them <strong>in</strong> some detail (i.e. AR(1)-GARCH(p, q) ×(P, Q) s , RCA(1)-GARCH(p, q) × (P, Q) s <strong>and</strong> RCA(1)-seasonal SV).<strong>The</strong> results are primarily oriented to f<strong>in</strong>ancial time series applications. F<strong>in</strong>ancial time series<strong>of</strong>ten meet the large dataset dem<strong>and</strong>s <strong>of</strong> the volatility models studied here. Also, f<strong>in</strong>ancial datadynamics <strong>in</strong> higher order moments are <strong>of</strong> <strong>in</strong>terest to many market participants. Specifically,we consider the Black-Scholes model with seasonal GARCH volatility <strong>and</strong> show that themoments <strong>of</strong> the seasonal volatility process can be used to evaluate the call price for Europeanoptions.8. ReferencesBadescu, A. & Kulperger, R. (2008). GARCH option pric<strong>in</strong>g: A semiparametric approach.Insurance, <strong>Mathematics</strong> & Economics, Vol. 43, No. 1, 69 – 84, ISSN 0167-6687
44 Advances <strong>in</strong> Econometrics - <strong>The</strong>ory <strong>and</strong> Applications14 Will-be-set-by-IN-TECHBaillie, R. & Bollerslev, T. (1990). Intra-day <strong>and</strong> <strong>in</strong>ter-market volatility <strong>in</strong> foreign exchangerates. Review <strong>of</strong> Economic Studies, Vol. 58, No. 3, 565 – 585, ISSN 0034-6527Bollerslev, T. (1986). Generalized autoregressive conditional heteroscedasticity. Journal <strong>of</strong>Econometrics, Vol. 31, 307 – 327, ISSN 0304-4076Bollerslev, T. & Ghysels, E. (1996). Periodic autoregressive conditional heteroscedasticity.Journal <strong>of</strong> Bus<strong>in</strong>ess <strong>and</strong> Economic Statistics, Vol. 14, No. 2, 139 – 151, ISSN: 0735-0015Doshi, A.; Frank, J.; Thavaneswaran, A. (2011). Seasonal volatility models. Journal <strong>of</strong> Statistical<strong>The</strong>ory <strong>and</strong> Applications, Vol. 10, No. 1, 1–10, ISSN 1538-7887Duan, J. (1995). <strong>The</strong> GARCH option ric<strong>in</strong>g model. Mathematical F<strong>in</strong>ance, Vol. 5, 13–32, ISSN0960-1627Elliot, R.; Siu, T.; Chan, L. (2006). Option pric<strong>in</strong>g for GARCH models with Markov switch<strong>in</strong>g.International Journal <strong>of</strong> <strong>The</strong>oretical <strong>and</strong> Applied F<strong>in</strong>ance, Vol. 9, No. 6, 825–841, ISSN0219-0249Engle, R. (1982). Autoregressive conditional heteroskedasticity with estimates <strong>of</strong> the variance<strong>of</strong> U.K. <strong>in</strong>flation. Econometrica, Vol. 50, 987–1008, ISSN 0012-9682Frank, J. & Garcia, P. (2009). Time-vary<strong>in</strong>g risk premium: further evidence <strong>in</strong> agriculturalfutures markets. Applied Economics, Vol. 41, No. 6, 715 – 725, ISSN 0003-6846Franses, P. & Paap, R. (2000). Modell<strong>in</strong>g day-<strong>of</strong>-the-week seasonality <strong>in</strong> the S&P 500 <strong>in</strong>dex.Applied F<strong>in</strong>ancial Economics, Vol. 10, No. 5, 483 – 488, ISSN 0960-3107Ghahramani, M. & Thavaneswaran, A. (2007). Identification <strong>of</strong> ARMA models with GARCHerrors. <strong>The</strong> Mathematical Scientist, Vol. 32, No. 1, 60 – 69, ISSN 0312-3685Ghysels, E. & Osborn, D. (2001). <strong>The</strong> econometric analysis <strong>of</strong> seasonal time series, CambridgeUniversity Press, ISBN 0-521-56588-X, United K<strong>in</strong>gdom.Gong, H. & Thavaneswaran, A. (2009). Recursive estimation for cont<strong>in</strong>uous time stochasticvolatility models. Applied <strong>Mathematics</strong> Letters, Vol. 22, No. 11, 1770 – 1774, ISSN0893-9659Gong, H.; Thavaneswaran, A.; S<strong>in</strong>gh, J. (2010). A Black-Scholes model with GARCH volatility.<strong>The</strong> Mathematical Scientist, Vol. 35, No. 1, 37–42, ISSN 0312-3685Heston, S. & N<strong>and</strong>i, S. (2000). A closed-form GARCH option valuation model. <strong>The</strong> Review <strong>of</strong>F<strong>in</strong>ancial Studies, Vol. 13, No. 3, 585 – 625, ISSN 0893-9454Nicholls, D. & Qu<strong>in</strong>n, B. (1982). R<strong>and</strong>om coefficient autoregressive models: An <strong>in</strong>troduction.Lecture Notes <strong>in</strong> Statistics, Vol. 11, Spr<strong>in</strong>ger, ISSN 0930-0325, New York.Paseka, A.; Appadoo, S.; Thavaneswaran, A. (2010). R<strong>and</strong>om coefficient autoregressive (RCA)models with nonl<strong>in</strong>ear stochastic volatility <strong>in</strong>novations. Journal <strong>of</strong> Applied StatisticalScience, Vol. 17, No. 3, 331–349, ISSN 1067-5817Taylor, S. (2005). Asset price dynamics, volatility, <strong>and</strong> prediction, Pr<strong>in</strong>ceton University Press, ISBN0-691-11537-0, New Jersey.Thavaneswaran, A. & Ghahramani, M. (2008). Volatility forecasts with GARCH errors <strong>and</strong>applications. Journal <strong>of</strong> Statistical <strong>The</strong>ory <strong>and</strong> Applications, Vol. 1, No. 1, 69 – 80, ISSN1538-7887Thavaneswaran, A.; Appadoo, S.; Ghahramani, M. (2009). RCA models with GARCH<strong>in</strong>novations. Applied <strong>Mathematics</strong> Letters, Vol. 22, 110–114, ISSN 0893-9659Thavaneswaran, A.; Appadoo, S.; Peiris, S. (2005a). Forecast<strong>in</strong>g volatility. Statistics <strong>and</strong>Probability Letters, Vol. 75, No. 1, 1–10, ISSN: 0167-7152Thavaneswaran, A.; Appadoo, S.; Samanta, M. (2005b). R<strong>and</strong>om coefficient GARCH Models.Mathematical <strong>and</strong> Computer Modell<strong>in</strong>g, Vol. 41, No. 1–7, 723–733, ISSN 0895-7177