Abstract
An important research area in statistical computing is found in the literature for vector functional coefficient autoregressive models, a special case of vector nonlinear time series. Methods used are computationally intensive. As a result, analyses and simulations can run into weeks, or even months. Statisticians have been known to base empirical results on a relatively small number of simulation replications, sacrificing precision, accuracy and reliability of results in the interest of time and productivity. The simulations are amenable for parallelization; however, parallel computing technology has not yet been widely used in this specific research area. This paper proposes an approach to the parallelization of statistical simulation codes to address the challenge of long running times, without resorting to extensive code revisions. This approach takes advantage of recent advances in dynamic loop scheduling on workstation clusters to achieve high performance, even with the presence of unpredictable load imbalance factors. Preliminary results of applying this approach in the simulation of normal white noise and threshold autoregressive model obtains efficiencies in the range 95–98% on 8–64 processors.
This work was partially supported by the National Science Foundation Grants: CAREER #9984465, ITR/DMS #0313274, ITR/ACS #0081303, ITR/ACS #0085969, #0132618, and #0082979.
Chapter PDF
Similar content being viewed by others
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
References
Gentle, J.: Random Number Generation and Monte Carlo Methods, 2nd edn. Springer, New York (2003)
Harvill, J., Ray, B.: Functional coefficient autoregressive models for vector time series. Computational Statistics and Data Analysis (2005) (To appear)
Harvill, J., Ray, B.: A note on multi-step forecasting with functional coefficient autoregressive models. International Journal of Forecasting (2005) (To appear)
Cariño, R.L., Banicescu, I.: A load balancing tool for distributed parallel loops. In: Proc. Int. Workshop on Challenges of Large Applications in Distributed Environments, pp. 39–46. IEEE Computer Society Press, Los Alamitos (2003)
Cariño, R.L., Banicescu, I.: Dynamic scheduling parallel loops with variable iterate execution times. In: Proc. 16th IEEE Int. Parallel and Distributed Processing Symposium - PDSECA, IEEE Computer Society Press, Los Alamitos (2002), CDROM
Cariño, R.L., Banicescu, I.: Load balancing parallel loops on message-passing systems. In: Akl, S., Gonzales, T. (eds.) Proc. 14th IASTED Int. Conf. on Parallel and Distributed Computing and Systems, pp. 362–367. ACTA Press (2002)
Cariño, R.L., Banicescu, I., Vadapalli, R.K., Weatherford, C.A., Zhu, J.: Message-passing parallel adaptive quantum trajectory method. In: Yang, L.T., Pan, Y. (eds.) High performance Scientific and Engineering Computing: Hardware/Software Support, pp. 127–139. Kluwer Academic Publishers, Dordrecht (2004)
Banicescu, I., Liu, Z.: Adaptive factoring: A dynamic scheduling method tuned to the rate of weight changes. In: Proc. High Performance Computing Symposium, pp. 122–129 (2000)
Banicescu, I., Velusamy, V.: Load balancing highly irregular computations with the adaptive factoring. In: Proc. 16th IEEE Int. Parallel and Distributed Processing Symposium - HCW, IEEE Computer Society Press, Los Alamitos (2002), CDROM
Banicescu, I., Velusamy, V., Devaprasad, J.: On the scalability of dynamic scheduling scientific applications with adaptive weighted factoring. Cluster Computing: The Journal of Networks, Software Tools and Applications 6, 215–226 (2003)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2005 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Banicescu, I., Cariño, R.L., Harvill, J.L., Lestrade, J.P. (2005). Computational Challenges in Vector Functional Coefficient Autoregressive Models. In: Sunderam, V.S., van Albada, G.D., Sloot, P.M.A., Dongarra, J.J. (eds) Computational Science – ICCS 2005. ICCS 2005. Lecture Notes in Computer Science, vol 3514. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11428831_30
Download citation
DOI: https://doi.org/10.1007/11428831_30
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-26032-5
Online ISBN: 978-3-540-32111-8
eBook Packages: Computer ScienceComputer Science (R0)