A Commentary on 'A Survey of Time Series'

Summary This commentary discusses a recent survey paper (concerned with time-series methodology) aiming to amend and to complement it. The present writer's admitted bias towards the time domain in general, and Box-Jenkins in particular, will be apparent - but that does reflect a strong current trend. Preface Spyros Makridakis (1976) presented an interesting and ambitious survey of time-series analysis, together with an impressive listing of 139 key references - which, perhaps with some modesty, included only three to work carrying his own name. The intention of that important 42-page paper was to look at the 'methods developed and used in the fields of statistics, engineering, operational research and economics'. For this purpose, it described 'each of the major methods now in existence' and synthesized 'them into an integrated whole by pointing out similarities and differences existing amongst them'. The applicability of the different methods to various real-life requirements was discussed and further research needs were highlighted. Evidently this was an extremely formidable task for any individual to undertake,2 and it is not surprising that another time-series analyst finds some fault with it. This article will comment on what appear to be certain errors and misprints; omissions and inadequacies of Makridakis' paper. However, these efforts must not be construed merely as a censorious contribution: had the present writer attempted such a wide-ranging review, he feels sure that Makridakis would have found much more to censure. This commentary is intended for those who have already read Makridakis (1976)- and were perhaps puzzled by parts of it - or at least have that paper to hand. Our subsequent sections and subsections will correspond exactly to those of his and we will endeavour to distinguish in this account between matters of fact, faith and fancy. (Fantasy, we hope, will be excluded!) With less modesty than Makridakis, I will cite more of my own works than their intrinsic worth warrants. This is because one naturally knows one's own material, and it more readily

[1]  W. Beveridge,et al.  Wheat Prices and Rainfall in Western Europe , 1922 .

[2]  Eugen Slutzky.  Summation of random causes as the source of cyclic processes , 1937 .

[3]  M. G. Kendall,et al.  The advanced theory of statistics. Vols. 2. , 1969 .

[4]  M. Kendall,et al.  A Study in the Analysis of Stationary Time-Series. , 1955 .

[5]  R. Meyer,et al.  The Fundamental Theorem of Exponential Smoothing , 1961 .

[6]  M. Kendall,et al.  The advanced theory of statistics , 1945 .

[7]  Robert Goodell Brown,et al.  Smoothing, forecasting and prediction of discrete time series , 1964 .

[8]  D. Cox,et al.  An Analysis of Transformations , 1964 .

[9]  D. G. Watts,et al.  Spectral analysis and its applications , 1968 .

[10]  Andrew H. Jazwinski,et al.  Adaptive filtering , 1969, Autom..

[11]  K. Wallis.  Some Recent Developments in Applied Econometrics: Dynamic Models and Simultaneous Equation Systems , 1969 .

[12]  Gwilym M. Jenkins,et al.  Time series analysis, forecasting and control , 1972 .

[13]  G. Box,et al.  Distribution of Residual Autocorrelations in Autoregressive-Integrated Moving Average Time Series Models , 1970 .

[14]  K. Åström.  Introduction to Stochastic Control Theory , 1970 .

[15]  Clive W. J. Granger,et al.  A New Look at Some Old Data: The Beveridge Wheat Price Series , 1971 .

[16]  G. Treyz,et al.  An Analysis of the Forecasting Properties of U.S. Econometric Models , 1972 .

[17]  Thomas H. Naylor,et al.  Box-Jenkins Methods: An Alternative to Econometric Models , 1972 .

[18]  R. L. Cooper.  The Predictive Performance of Quarterly Econometric Models of the United States , 1972 .

[19]  William S. Cleveland,et al.  The Inverse Autocorrelations of a Time Series and Their Applications , 1972 .

[20]  George E. P. Box,et al.  Some Comments on a Paper by Chatfield and Prothero and on a Review by Kendall , 1973 .

[21]  J. Dickinson.  Some Statistical Results in the Combination of Forecasts , 1973 .

[22]  C. Chatfield,et al.  Box‐Jenkins Seasonal Forecasting: Problems in a Case‐Study , 1973 .

[23]  Spyros Makridakis,et al.  An Examination of the Use of Adaptive Filtering in Forecasting , 1973 .

[24]  B. Hickman,et al.  Econometric models of cyclical behavior , 1973 .

[25]  L. H. Koopmans.  The spectral analysis of time series , 1974 .

[26]  M. J. Murphy,et al.  Seasonal Adjustment Based on a Mixed Additive‐Multiplicative Model , 1975 .

[27]  George E. P. Box,et al.  Comparison of Forecasts and Actuality. , 1975 .

[28]  D. Hester,et al.  Modelling the Economy. , 1975 .

[29]  Chris Chatfield.  The Analysis of Time Series: Theory and Practice , 1975 .

[30]  O. Anderson.  Bounding sums for the autocorrelations of moving average processes , 1975 .

[31]  D. A. Pierce.  On trend and autocorrelation , 1975 .

[32]  George E. P. Box,et al.  Intervention Analysis with Applications to Economic and Environmental Problems , 1975 .

[33]  O. D. Anderson,et al.  On a lemma associated with Box, Jenkins and Granger , 1975 .

[34]  O. Anderson.  Moving Average Processes , 1975 .

[35]  Paul Newbold,et al.  The Principles of the Box-Jenkins Approach , 1975 .

[36]  O. D. Anderson,et al.  Distinguishing between simple Box—Jenkins models , 1975 .

[37]  O. D. Anderson,et al.  Time Series Analysis and Forecasting: The Box-Jenkins Approach. , 1976 .

[38]  O. D. Anderson,et al.  The backshift operator in time series analysis , 1976 .

[39]  Letter: On Bulmer's statistical analysis of density dependence. , 1976 .

[40]  Steinar Ekern.  Forecasting with Adaptive Filtering: A Critical Re-examination , 1976 .

[41]  A. Ehrenberg.  We must preach what is practised: A radical review of statistical teaching , 1976 .

[42]  Ed McKenzie.  A Comparison of Some Standard Seasonal Forecasting Systems , 1976 .

[43]  G. Tunnicliffe Wilson,et al.  Interpolating Time Series with Application to the Estimation of Holiday Effects on Electricity Demand , 1976 .

[44]  E. R. Golder,et al.  On Adaptive Filtering , 1976 .

[45]  Kenneth F. Wallis,et al.  Modelling Macroeconomic Time Series , 1976 .

[46]  Dynamic Modeling of Multivariate Time Series for Use in Bank Analysis , 1976 .

[47]  C. Granger,et al.  Forecasting transformed series , 1976 .

[48]  Spyros Makridakis.  A Survey of Time Series , 1976 .

[49]  O. D. Anderson,et al.  The Interpretation of Box-Jenkins Time Series Models , 1977 .

[50]  O. Anderson.  An inequality and a lemma revisited , 1977 .

[51]  O. Anderson.  On sorting out Poole's paper "Stochastic difference equation predictors of population fluctuations" about the Box-Jenkins analysis and forecasting of ecological time series. , 1978, Theoretical population biology.