Abstract
This paper proposes a modified approach to the combination of forecasts from multivariate volatility models where the combination is performed over a restricted subset including only the best performing models. Such a subset is identified over a rolling window by means of the Model Confidence Set (MCS) approach. The analysis is performed using different combination schemes, both linear and non linear, and considering different loss functions for the evaluation of the forecasting performance. An application to a vast dimensional portfolio of 50 NYSE stocks shows that (a) in non-extreme volatility periods the use of forecast combinations allows to improve over the predictive accuracy of the single candidate models (b) performing the combination over the subset of most accurate models does not significantly reduce the accuracy of the combined predictor.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The data are available online at www.tickdata.com.
- 2.
The choice of a discretization interval equal to 5Â min is a standard practice in the literature since it usually guarantees a reasonable compromise between bias and variability of the realized covariance estimator.
References
Amendola, A., Storti, G.: A GMM procedure for combining volatility forecasts. Comput. Stat. Data Anal. 52(6), 3047–3060 (2008)
Amendola, A., Storti, G.: Combination of multivariate volatility forecasts. SFB 649 Discussion Papers, DP2009-007. Humboldt University, Berlin, Germany (2009)
Amendola, A., Storti, G.: Model uncertainty and forecast combination in high dimensional multivariate volatility prediction. In: Proceedings of COMPSTAT 2012, ISI/IASC, 27–38 (2012)
Andersen, T.G., Bollerslev, T., Frederiksen, P., Nielsen, O.: Continuous-time models, realized volatilities, and testable distributional implications for daily stock returns. J. Appl. Econometrics 25(2), 233–261 (2010)
De Pooter, M., Ravazzolo, F., van Dijk, D.: Term structure forecasting using macro factors and forecast combination. Board of Governors of the Federal Reserve System, Discussion Paper 993 (2010)
Engle, R.F.: Dynamic conditional correlation: a simple class of multivariate generalized autoregressive conditional heteroskedasticity models. J. Bus. Econ. Stat. 20(3), 339–350 (2002)
Engle, R.F., Shephard, N., Sheppard, K.: Fitting vast dimensional time-varying covariance models. Economics Series Working Papers 403. University of Oxford, Oxford (2008)
Engle, R.F., Kroner, K.F.: Modelling the coherence in short-run nominal exchange rates: a multivariate generalized arch model. Econ. Theor. 11(1), 122–150 (1995)
Golosnoy, V., Gribisch, B., Liesenfeld, R.: The conditional autoregressive Wishart model for multivariate stock market volatility. J. Econometrics 167, 211–223 (2011)
Granger, C.W.J., Jeon, Y.: Thick modeling. Econ. Model. 21, 323–343 (2004)
Hansen, P.R., Lunde, A., Nason, J.M.: The model confidence set. Econometrica 79, 453–497 (2011)
J.P. Morgan Guaranty Trust Company: RiskMetrics Technical Document, 4 edn. (1996)
Laurent, S., Rombouts, J.V.K., Violante, F.: On loss functions and ranking forecasting performances of multivariate volatility models. J. Econometrics 173(1), 1–10 (2013)
Patton, A.J., Sheppard, K.: Evaluating volatility and correlation forecasts. In: Andersen, T.G., Davis, R.A., Kreiss, J.P., Mikosch, T. (eds.) Handbook of Financial Time Series. Springer, Berlin (2009)
Patton, A.J.: Volatility forecast comparison using imperfect volatility proxies. J. Econometrics 160(1), 246–256 (2011)
Pesaran, M.H., Schleicher, C., Zaffaroni, P.: Model averaging in risk management with an application to futures markets. J. Empir. Finance 16(2), 280–305 (2009)
Acknowledgements
The authors gratefully acknowledge financial support from MIUR within the PRIN project 2010–2011 (prot. 2010J3LZEN): Forecasting economic and financial time series: understanding the complexity and modelling structural change.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Amendola, A., Storti, G. (2014). A Thick Modeling Approach to Multivariate Volatility Prediction. In: Carpita, M., Brentari, E., Qannari, E. (eds) Advances in Latent Variables. Studies in Theoretical and Applied Statistics(). Springer, Cham. https://doi.org/10.1007/10104_2014_18
Download citation
DOI: https://doi.org/10.1007/10104_2014_18
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-02966-5
Online ISBN: 978-3-319-02967-2
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)