A Conversation with Hirotugu Akaike

  • David F. Findley
  • Emanuel Parzen
Part of the Springer Series in Statistics book series (SSS)

Abstract

Hirotugu Akaike was born in Fujinomiya City, Shizuoka Prefecture, Japan on the fifth of November 1927. He studied at the Naval Academy of Japan, the First Higher School and the University of Tokyo, where he earned his B.S. degree and his external Doctor of Science degree, both in mathematics.

After receiving his bachelor’s degree in 1952, he was hired by the Institute of Statistical Mathematics, which had been founded eight years earlier by the Japanese government. He was Director of the institute’s Fifth Division, concerned with time series analysis and control, from 1973 until 1985. When the institute was reorganized as an interuniversity research institute in 1986, he became a Professor and Director of the Department of Prediction and Control. In 1987, he became Director General of the Institute, the position from which he retired on March 31, 1994. He was also Professor and Head of the Department of Statistical Science of the Graduate University for Advanced Studies, an independent university whose departments are distributed among the 11 interuniversity research institutes, from 1988 until 1994.

He has held visiting positions at a number of universities: Princeton (1966–1967), Stanford (1967, 1979), Hawaii (1972), the University of Manchester Institute of Science and Technology (1973), Harvard (Vinton Hayes Senior Fellow in Engineering and Applied Physics, 1976), Wisconsin—Madison (Mathematics Research Center, 1982) and several Japanese universities.

His honors include two major technology prizes, each shared with one or more collaborating engineers: with Toichiro Nakagawa, he was awarded the 1972 Ishikawa Prize for modernization of production management by the Ishikawa Prize Committee of the Japan Union of Scientists and Engineers; and, with Hideo Nakamura and others, he received the 1980 Okochi Prize of the Okochi Memorial Foundation for contributions to production engineering. In 1989, he was the recipient of two of Japan’s most respected culture and science awards, the Purple Ribbon Medal given by the Emperor of Japan and the Asahi Prize of the Asahi Shimbun Foundation, awards which recognize writers and artists and other citizens as well as inventors and scientists for distinguished contributions to Japanese society. He was a member of the Science Council of Japan from 1988 to 1991.

He has published more than 140 papers and several monographs and textbooks. His 1972 monograph with T. Nakagawa on the statistical analysis and control of dynamic systems has been republished in English translation (Akaike and Nakagawa, 1988). To indicate the magnitude of the impact of the methods described in this book, Professor Genshiro Kitagawa kindly provided us with a table from an article published in Japan in February 1994 listing the outputs of electric power plants in Japan that were built to be controlled by statistical models based on these methods. The table shows these plants generated approximately 12% of Japan’s electrical power obtained from nonnuclear and nonhydroelectric sources.

The initial conversation, in which David Findley and Emanuel Parzen spoke with Professor Akaike, took place in May 1992 at the University of Tennessee in Knoxville during the “First U.S.—Japan Conference on the Frontiers of Statistical Modeling: An Information Approach.” Findley later obtained clarifications and amplifications of some points from Professor Akaike during visits to the Institute of Statistical Mathematics in Tokyo in March 1993 and February 1994.

Keywords

Entropy Covariance Transportation Shrinkage Bicarbonate 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Araire, H. (1956). On a zero-one process and some of its applications. Ann. Inst. Statist. Math. 8 87–94.MathSciNetCrossRefGoogle Scholar
  2. Araire, H. (1959). On the statistical control of the gap process. Ann. Inst. Statist. Math. 10 233–259.MathSciNetCrossRefGoogle Scholar
  3. Araire, H. (1967). Some problems in the application of the cross spectral method. In Spectral Analysis of Time Series ( B. Harris, ed.) 81–107. Wiley, New York.Google Scholar
  4. Araire, H. (1968). On the use of a linear model for the identification of feedback systems. Ann. Inst. Statist. Math. 20 425–429.MathSciNetCrossRefGoogle Scholar
  5. Araire, H. (1970). On a semi-automatic power spectrum estimation procedure. In Proceedings of the Third Hawaii International Conference on System Sciences 974–977. Western Periodicals Company, Honolulu.Google Scholar
  6. Akaike, H. (1974a). A new look at the statistical model identifi- cation. IEEE Trans. Automat. Control AC-19 716–723.MathSciNetCrossRefGoogle Scholar
  7. Araire, H. (1974b). Markovian representation of stochastic processes and its application to the analysis of autoregressive moving average processes. Ann. Inst. Statist. Math. 26 363–387.MathSciNetCrossRefGoogle Scholar
  8. Akaike, H. (1977a). An objective use of Bayesian models. Ann. Inst. Statist. Math. 29 9–20.MathSciNetCrossRefMATHGoogle Scholar
  9. Araire, H. (1977b). On entropy maximization principle. In Applications of Statistics (P. R. Krishnaiah, ed.) 27–41. North-Holland, Amsterdam.Google Scholar
  10. Araire, H. (1977c). Canonical correlation analysis of time series and the use of an information criterion. In System Identification: Advances and Case Studies (R. K. Mehra and D. G. Laniotis, eds.) 29–76. Academic, New York.Google Scholar
  11. Araire, H. (1978). A Bayesian analysis of the minimum BIC procedure. Ann. Inst. Statist. Math. 30 9–14.MathSciNetCrossRefGoogle Scholar
  12. Araire, H. (1979). A Bayesian extension of the minimum AIC procedure of autoregressive model fitting. Biometrika 66 237–242.MathSciNetCrossRefGoogle Scholar
  13. Araire, H. (1980). Likelihood and the Bayes procedure. In Bayesian Statistics (J. M. Bernardo, M. H. DeGroot, D. V. Lindley and A. F. M. Smith, eds.) 143–166. Univ. Press Valencia.Google Scholar
  14. Araire, H. (1994). Implications of informational point of view on the development of statistical science. In Proceedings of the First US/JAPAN Conference on The Frontiers of Statistical Modeling: An Informational Approach 3 (H. Bozdogan, ed.) 27–38. Kluwer, Dordrecht.Google Scholar
  15. Araire, H. and Knceshioe, I. (1962). Some estimation of vehicle suspension system’s frequency response by cross-spectral method. In Proceedings of the 12th Japan National Congress for Applied Mechanics 241–244. Univ. Tokyo Press.Google Scholar
  16. Araire, H. and Nakagawa, T. (1988). Statistical Analysis and Control of Dynamic Systems. Kluwer, Dordrecht.Google Scholar
  17. Akaike, H. and Yamanouchi, Y. (1962). On the statistical estimation of frequency response function. Ann. Inst. Statist. Math. 14 23–56.MathSciNetCrossRefMATHGoogle Scholar
  18. Davisson, L. D. (1966). The prediction error of stationary Gaussian time series of unknown covariance. IEEE Trans. Inform. Theory IT-11 527–532.Google Scholar
  19. Fisher, R. A. (1935). The logic of inductive inference. J. Roy. Statist. Soc. Ser. A 98 39–54.CrossRefGoogle Scholar
  20. Fuiunishi, K. (1977). Diagnostic analyses of a nuclear power plant using multivariate autoregressive processes. Nuclear Science and Engineering 62 215–225.Google Scholar
  21. Johnson, N. L. and Kotz. S., eds. (1991). Breakthroughs in Statistics. Volume I: Foundations and Basic Theory. Springer, New York.Google Scholar
  22. Jones, R. H. (1975a). Probability estimation using a multinomial logistic function. J. Statist. Comput. Simulation 3 315–329.CrossRefMATHGoogle Scholar
  23. Jones, R. H. (1975b). Estimation of spatial wave number spectra and falloff rate with unequally spaced observations. J. Atmospheric Sci. 32 260–268.CrossRefGoogle Scholar
  24. Nakamura, H. (1994). Statistical identification and optimal control of thermal power plants. In Proceedings of the First US/Japan Conference on the Frontiers of Statistical Modeling: An Informational Approach 3 ( H. Bozdogan, ed.) 57–79. Kluwer, Dordrecht.Google Scholar
  25. Nakamura, H. and Araire, H. (1981). Statistical identification for optimal control of supercritical thermal power plants. Automatica 17 143–155.CrossRefMATHGoogle Scholar
  26. Peirce, C. S. (1955). Abduction and induction. In Philosphical Writings of Peirce (J. Buehler, ed.), 150–156. Dover, New York.Google Scholar
  27. This Week’s Citation Ciassic (1981). Current Contents 51 22 (December 21). [Also included in (1986) Contemporary Classics in Engineering and Applied Science 42 ( A. Thackary, ed.) ISI Press, Philadelphia.Google Scholar
  28. Tikhngv, A. N. (1965). Incorrect problems of linear algebra and a stable method for their solution Soviet Math. Dokl. 6 989–991.Google Scholar
  29. Wada, T., Axnluuue, H., Yamada, Y. and Uoagawa, E. (1988). Application of multivariate autoregressive modelling for analysis of immunologic networks in man. Comput. Math. Appt. 15 713–722.Google Scholar
  30. Wada, T., Sata, S. and Matsuo, N. (1993). Application of multivariate autoregressive modeling for analyzing chloride/ potassium/bicarbonate relationship in the body. Medical and Biological Engineering and Computers 31 S99 - S107.CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 1998

Authors and Affiliations

  • David F. Findley
  • Emanuel Parzen

There are no affiliations available

Personalised recommendations