KIEE International Transactions on Power Engineering

The Korean Institute of Electrical Engineers (대한전기학회)
권호 표지

Accounting for Extreme Values in GARCH Forecasts of Day-Ahead Electricity Prices

  • Guirguis Hany S. (Dept. of Economic and Finance, School of Business, Manhattan College) ;
  • Felder Frank A. (The State University of New Jersy)
  • Published : 2005.09.01
Download PDF
⟨ Previous Next ⟩

Abstract

We employ a new technique to account for extreme values when using the generalized autoregressive conditionally heteroskedastic (GARCH) methodology to forecast day-ahead electricity prices in New York City.

Keywords

References

  1. H. Bystrom, 'Extreme Value Theory and Extremely Large Electricity Price Changes,' Department of Economics, Lund University, Oct. 15, 2001, unpublished
  2. C. R. Knittel and M. R. Roberts, 'An Empirical Examination of Deregulated Electricity Prices,' PWP-087, University of California Energy Institute, Oct. 2001
  3. T. Bollerslev, 'Generalized autoregressive conditional heteroscedasticity,' Journal of Economics, 31, 1986, pp. 302-327
  4. NYISO Report, New York Power Pool Zone Forecasting Models, February 10, 1999, p. 1, available at nyiso.com
  5. E. Walter, 'Applied Econometric Time Series,' New York, NY: John Wiley & Sons, 1994, pp. 233-243 and pp. 146-148
  6. W. H. Press, B. P. Flannery, S. A. Teukolsky and W. T. Vettering Numerical Recipes in C, New York, NY: Cambridge University Press, 1988
  7. Franses, P. and H. Ghijsels, 'Additive outliers, GARCH and forecasting volatility,' International Journal of Forecasting 15, 1999, pp. 1-9 https://doi.org/10.1016/S0169-2070(98)00053-3
  8. Inclan, C. and G. C. Tiao, 'Use of cumulative sums of squares for retrospective detection of changes of variance,' Journal of the American Statistical Association 89, 1994, pp. 913-923 https://doi.org/10.2307/2290916