Abstract
This paper deals with simultaneous prediction for time series models. In particular, it presents a simple procedure which gives well-calibrated simultaneous prediction intervals with coverage probability close to the target nominal value. Although the exact computation of the proposed intervals is usually not feasible, an approximation can be easily attained by means of a suitable bootstrap simulation procedure. This new predictive solution is much simpler to compute than those ones already proposed in the literature, based on asymptotic calculations. Applications of the bootstrap calibrated procedure to AR, MA and ARCH models are presented.
Similar content being viewed by others
References
Alpuim MT (1997) One-sided simultaneous prediction intervals for AR(1) and MA(1) processes with exponential innovations. J Forecast 16:19–35
Arellano-Valle RB, Genton MG (2008) On the exact distribution of the maximum of absolutely continuous dependent random variables. Stat Probab Lett 78:27–35
Barndorff-Nielsen OE, Cox DR (1996) Prediction and asymptotics. Bernoulli 2:319–340
Beran R (1990) Calibrating prediction regions. J Am Stat Assoc 85:715–723
Clements MP, Kim JH (2007) Bootstrap prediction intervals for autoregressive time series. Comput Stat Data Anal 51:3580–3594
Corcuera JM, Giummolè F (2006) Multivariate prediction. Bernoulli 12:157–168
Fonseca G, Giummolè F, Vidoni P (2014) Calibrating predictive distributions. J Stat Comput Sim 84:373–383
Giummolè F, Vidoni P (2010) Improved prediction limits for a general class of Gaussian models. J Time Ser Anal 31:483–493
Hall P, Peng L, Tajvidi N (1999) On prediction intervals based on predictive likelihood or bootstrap methods. Biometrika 86:871–880
Kabaila P (1999) An efficient simulation method for the computation of a class of conditional expectations. Aust NZ J Stat 41:331–336
Kabaila P, Syuhada K (2007) Improved prediction limits for AR(p) and ARCH(p) processes. J Time Ser Anal 29:213–223
Kim J (1999) Asymptotic and bootstrap prediction regions for vector autoregressions. Int J Forecast 15:393–403
Nadarajah S, Kotz S (2008) Exact distribution of the Max/Min of two Gaussian random variables. IEEE Trans VLSI Syst 16:210–212
Nolan JP, Ravishanker N (2009) Simultaneous prediction intervals for ARMA processes with stable innovations. J Forecast 28:235–246
Ravishanker N, Wu LSY, Glaz J (1991) Multiple prediction intervals for time series: comparison of simultaneous and marginal intervals. J Forecast 10:445–463
Vidoni P (2004) Improved prediction intervals for stochastic process models. J Time Ser Anal 25:137–154
Wolf M, Wunderli D (2015) Bootstrap joint prediction regions. J Time Ser Anal 36:352–376
Acknowledgements
This research was partially supported by the University of Udine under the PRID2017 research grants and by the Italian Ministry for University and Research under the PRIN2015 Grant No. 2015EASZFS_003. Simulation studies were performed on SCSCF, a multiprocessor cluster system owned by Università Ca’ Foscari Venezia.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Fonseca, G., Giummolè, F. & Vidoni, P. A note on simultaneous calibrated prediction intervals for time series. Stat Methods Appl 30, 317–330 (2021). https://doi.org/10.1007/s10260-020-00526-6
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10260-020-00526-6