Summary
The purpose of this paper is not to answer a question, but to ask one. Why do time series analysts have to skilfully identify and efficiently estimate time series models? This is time consuming and computationally costly, and with the Box-Jenkins approach apparently requires much expertise and experience. A number of criticisms of the standard methodology are made, and some warnings given. Several practical suggestions are proposed; but the article is intended to be constructive and in no way destructive of a very fine technique. It is felt that difficulties and dangers, honestly acknowledged, are that much the easier to overcome or avoid, respectively.
This is a preview of subscription content, log in via an institution to check access.
References
Anderson, O. D., ‘An Inequality with a Time Series Application,’Journal of Econometrics, II (1974), pp. 189–193.
Anderson, O. D.,Time Series Analysis and Forecasting: The Box-Jenkins Approach, London and Boston, 1975.
Anderson, O. D., ‘Bounding Sums for the Autocorrelations of Moving Average Processes,’Biometrika, LXII (1975), pp. 706–707.
Anderson, O. D., ‘Moving Average Processes,’The Statistician, XXIV (1975), pp. 283–297.
Anderson, O. D., ‘Distinguishing between simple Box-Jenkins models,’International Journal of Mathematical Education in Science and Technology, VI (1975), pp. 461–465. (This material was originally presented at the 1974 Annual Conference of the Institute of Statisticians, York.)
Anderson, O. D., ‘Box-Jenkins in Government: a development in official forecasting,’Statistical News, XXXII (1976), pp. 14–20.
Box, G. E. P. and G. M. Jenkins,Time Series Analysis forecasting and control, San Francisco, 1970.
Chatfield, C. and D. L. Prothero, ‘Box-Jenkins Seasonal Forecasting: Problems in a Case-Study (with Discussion, some Comments from Box and Jenkins, and a reply to these),’Journal of the Royal Statistical Society, Series A, CXXXVI (1973), pp. 295–352.
Harrison, P. J. and C. F. Stevens, ‘Bayesian Forecasting,’ (with Discussion),Journal of the Royal Statistical Society, Series B, XXXVIII (1976), pp. 205–247.
I C L,1 9 0 0 Series Technical Publication, London, 1972.
Jenkins, G. M., ‘The Interaction between the Muskrat and Mink cycles in North Canada,’Proceedings of the eighth International Biometrical Conference (1975), pp. 55–71.
Kendall, M. G., Book Review,Journal of the Royal Statistical Society, Series A, CXXXIV (1971), pp. 450–453.
Nelson, Charles R.,Applied Time Series Analysis for Managerial Forecasting, San Francisco, 1973.
Newbold, P., ‘The principles of the Box-Jenkins approach,’Operational Research Quarterly, XXVI (1975), pp. 397–412. (First presented at Conference on Practical Aspects of Forecasting, Imperial College, London, 1973. Proceedings published 1975, Edited by H. A. Gordon).
Stern, G. J. A., In Discussion of [8], pp. 329–330.
Tomasek, O., ‘Statistical Forecasting of Telephone Time Series,’ITO Telecommunications Journal, (December 1972), pp. 1–7.
Additional information
This paper was written while the author was with the Department of Mathematics, Nottingham University. Currently, Dr. Anderson is a government statistician and course director at the London centre of the Civil Service College. Present address 11 Belgrave Road, London SW 1 V 1RB, England.
Rights and permissions
About this article
Cite this article
Anderson, O.D. Is Box-Jenskins a waste of time?. De Economist 125, 254–263 (1977). https://doi.org/10.1007/BF01225612
Issue Date:
DOI: https://doi.org/10.1007/BF01225612