• Anderson, T. W., 1984: An Introduction to Multivariate Statistical Analysis. 2d ed. Series in Probability and Mathematical Statistics, Wiley, 675 pp.

    • Search Google Scholar
    • Export Citation
  • Barnett, T. P., and R. Preisendorfer, 1987: Origins and levels of monthly and seasonal forecast skill for United States surface air temperatures determined by canonical correlation analysis. Mon. Wea. Rev., 115 , 18251850.

    • Search Google Scholar
    • Export Citation
  • Branstator, G., A. Mai, and D. Baumhefner, 1993: Identification of highly predictable flow elements for spatial filtering of medium- and extended-range numerical forecasts. Mon. Wea. Rev., 121 , 17861802.

    • Search Google Scholar
    • Export Citation
  • DelSole, T., 2002: Entropy as a basis for comparing and blending forecasts. Quart. J. Roy. Meteor. Soc., in press.

  • Déqué, M., 1988: 10-day predictability of the Northern Hemisphere winter 500mb height by the ECMWF operational model. Tellus, 40A , 2636.

    • Search Google Scholar
    • Export Citation
  • Newman, M., and P. D. Sardeshmukh, 1995: A caveat concerning singular value decomposition. J. Climate, 8 , 352360.

  • Noble, B., and J. W. Daniel, 1988: Applied Linear Algebra. 3d ed. Prentice-Hall, 521 pp.

  • Penland, C., and P. D. Sardeshmukh, 1995: The optimal growth of tropical sea surface temperature anomalies. J. Climate, 8 , 19992024.

  • Renwick, J. A., and J. M. Wallace, 1995: Predictable anomaly patterns and the forecast skill of Northern Hemisphere wintertime 500-mb height fields. Mon. Wea. Rev., 123 , 21142131.

    • Search Google Scholar
    • Export Citation
  • Schneider, T., and S. M. Griffies, 1999: A conceptual framework for predictability studies. J. Climate, 12 , 31333155.

  • Thacker, W. C., 1996: Metric-based principal components: Data uncertainties. Tellus, 48A , 584592.

  • von Storch, H., and F. W. Zwiers, 1999: Statistical Analysis in Climate Research. Cambridge University Press, 484 pp.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 346 118 6
PDF Downloads 221 92 7

Predictable Component Analysis, Canonical Correlation Analysis, and Autoregressive Models

Timothy DelSoleCenter for Ocean–Land–Atmosphere Studies, Calverton, Maryland

Search for other papers by Timothy DelSole in
Current site
Google Scholar
PubMed
Close
and
Ping ChangDepartment of Oceanography, Texas A&M University, College Station, Texas

Search for other papers by Ping Chang in
Current site
Google Scholar
PubMed
Close
Restricted access

Abstract

The purpose of this paper is to clarify the relation between canonical correlation analysis, autoregressive models (also called linear inverse models), and a relatively new statistical technique known as predictable component analysis. Predictable component analysis is a procedure for decomposing forecast error into a complete set of uncorrelated patterns that optimize the normalized error variance. As such, the procedure determines the components of a forecast, out of all possible components, that are predicted the best and worst on average, in a normalized error variance sense. This procedure has been suggested previously in the context of information theory and of maximizing the variance of prewhitened forecast errors. It is shown that the most predictable components of a linear, first-order, autoregressive model are identical to the canonical patterns that optimize the temporal correlation between the original and time-lagged time series. Also, canonical patterns define a transformation that diagonalizes the propagator of the autoregressive model. Finally, the singular vectors of an autoregressive propagator have a one-to-one correspondence to the canonical patterns, provided that the singular vectors are computed with respect to a norm related to a prewhitening transformation. Forecasts based on canonical correlation analysis are shown to be identical to forecasts based on autoregressive models when all canonical patterns are superposed. The advantage of canonical correlation analysis is that it allows one to filter out the “unpredictable components” of autoregressive models, defined as components in the forecast with skill below some statistical threshold of significance. Predictable component analysis allows a generalization of this filtering procedure to any forecast model.

Corresponding author address: Timothy M. DelSole, 4041 Powder Mill Rd., Suite 302, Calverton, MD 20705-3106. Email: delsole@cola.iges.org

Abstract

The purpose of this paper is to clarify the relation between canonical correlation analysis, autoregressive models (also called linear inverse models), and a relatively new statistical technique known as predictable component analysis. Predictable component analysis is a procedure for decomposing forecast error into a complete set of uncorrelated patterns that optimize the normalized error variance. As such, the procedure determines the components of a forecast, out of all possible components, that are predicted the best and worst on average, in a normalized error variance sense. This procedure has been suggested previously in the context of information theory and of maximizing the variance of prewhitened forecast errors. It is shown that the most predictable components of a linear, first-order, autoregressive model are identical to the canonical patterns that optimize the temporal correlation between the original and time-lagged time series. Also, canonical patterns define a transformation that diagonalizes the propagator of the autoregressive model. Finally, the singular vectors of an autoregressive propagator have a one-to-one correspondence to the canonical patterns, provided that the singular vectors are computed with respect to a norm related to a prewhitening transformation. Forecasts based on canonical correlation analysis are shown to be identical to forecasts based on autoregressive models when all canonical patterns are superposed. The advantage of canonical correlation analysis is that it allows one to filter out the “unpredictable components” of autoregressive models, defined as components in the forecast with skill below some statistical threshold of significance. Predictable component analysis allows a generalization of this filtering procedure to any forecast model.

Corresponding author address: Timothy M. DelSole, 4041 Powder Mill Rd., Suite 302, Calverton, MD 20705-3106. Email: delsole@cola.iges.org

Save