Summary

Proceedings of the 2013 International Symposium on Nonlinear Theory and its Applications

2013

Session Number:B3L-A

Session:

Number:290

Probabilistic Forecast Combination by Geometric Averages

Devin Kilminster,  

pp.290-293

Publication Date:

Online ISSN:2188-5079

DOI:10.15248/proc.2.290

PDF download (362.6KB)

Summary:
We show how the Principle of Maximum Entropy suggests a method for probabilistic forecast combination involving a normalised weighted geometric average of the component probability density functions. An application to operational weather forecasting will be described and some results demonstrating the performance of the method will be presented.

References:

[1] J. M. Bernado. Expected information as expected utility. Annals of Statistics, 7, 1979.

[2] B. Efron and R. Tibshirani. Bootstrap methods for standard errors, confidence intervals and other measures of statistical accuracy. Statistical Science, 1:54-77, 1986.

[3] H. R. Glahn and D. A. Lowry. The use of model output statistics (MOS) in objective weather forecasting. J. Appl. Meteor., 11:1203-1211, 1972.

[4] I. J. Good. Rational decisions. Journal of the Royal Statistical Society: SERIES B (METHODOLOG ICAL), XIV(1):107-114, 1952.

[5] E. T. Jaynes. Information Theory and Statistical Mechanics. Physical Review, 106:620-630, May 1957.

[6] E. T. Jaynes. Probability Theory: The Logic of Science. Cambridge University Press, Apr. 2003.

[7] E. Kalnay. Atmospheric Modeling, Data Assimilation and Predictability. Cambridge University Press, 2003.

[8] M. S. Roulston and L. A. Smith. Evaluating probabilistic forecasts using information theory. Monthly Weather Review, 130(6):1653-1660, 2002.