Estimation of the Semiparametric Factor Model: Application to Modelling Time Series of Electricity Spot Prices.

Liebl, Dominik (2010): Estimation of the Semiparametric Factor Model: Application to Modelling Time Series of Electricity Spot Prices.

Abstract

Classical univariate and multivariate time series models have problems to deal with the high variability of hourly electricity spot prices. We propose to model alternatively the daily mean electricity supply functions using a dynamic factor model. And to derive, subsequently, the hourly electricity spot prices by the evaluation of the estimated supply functions at the corresponding hourly values of demand for electricity. Supply functions are price (EUR/MWh) functions, that increase monotonically with demand for electricity (MW). Apart from this new conceptual approach, that allows us to represent the auction design of energy exchanges in a most natural way, our main contribution is an extraordinary simple algorithm to estimate the factor structure of the dynamic factor model. We decompose the time series into a functional spherical component and an univariate scaling component. The elements of the spherical component are all standardized having unit size such that we can robustly estimate the factor structure. This algorithm is much simpler than procedures suggested in the literature. In order to use a parsimonious labeling we will refer to the daily mean supply curves simply as price curves.

• Item Type: MPRA Paper
• Language: English
• Keywords: Factor Analysis; functional time series data; sparse data; electricity spot market prices; European Electricity Exchange (EEX)
• Subjects: C - Mathematical and Quantitative Methods > C1 - Econometric and Statistical Methods and Methodology: General > C14 - Semiparametric and Nonparametric Methods: General; C - Mathematical and Quantitative Methods > C2 - Single Equation Models ; Single Variables > C22 - Time-Series Models ; Dynamic Quantile Regressions ; Dynamic Treatment Effect Models ; Diffusion Processes; C - Mathematical and Quantitative Methods > C1 - Econometric and Statistical Methods and Methodology: General; C - Mathematical and Quantitative Methods > C0 - General > C01 - Econometrics

References

J Antoch, L Prchal, MR Rosa, and P Sarda. Functional linear regression with functional response: Application to prediction of electricity consumption. Functional and Operatorial Statistics, pages 5219-5219, 2008.

M Forni, M Hallin, M Lippi, and M Reichlin. The generalized dynamic factor model: Identication and estimation. The Review of Economics and Statistics, (82): 540-554, 2000.

W K Härdle and S Trück. The dynamics of hourly electricity prices. 2010.

B U Park, E Mammen, W Härdle, and S Borak. Time Series Modelling With Semiparametric Factor Dynamics Journal of the American Statistical Association, 104(485): 284-298, 2009.

J G Staniswalis and J J Lee. Nonparametric Regression Analysis of Longitudinal Data. Journal of the American Statistical Association, 93(444): 1403-1418, 1998.

F Yao, H G Müller, and J L Wang. Functional Data Analysis for Sparse Longitudinal Data. Journal of the American Statistical Association, 100(470): 577-590, 2005.