A risk analysis method for carbon price prediction with hybrid intelligent model in consideration of variable selection of graphical modeling

This paper proposes a new risk assessment method for short-term carbon price prediction model. In this paper, a hybrid intelligent method of DA clustering and artificial neural network (ANN) is presented as a predictor of short-term carbon price. DA clustering plays a key role to classify input data into clusters. ANN is useful for predicting one-step ahead carbon price at each cluster. Graphical modeling is used to select meaningful input variables and provide more realistic relationship between input and output variables in the prediction model. To evaluate the uncertainty of the day-ahead carbon low, Monte-Carol simulation is carried out to generate sufficient realistic pseudo-scenarios with the multivariate normal random number. The proposed method is successfully applied to the real market data.

[1]  T. Dillon,et al.  Electricity price short-term forecasting using artificial neural networks , 1999 .

[2]  Nanny Wermuth,et al.  Algorithm AS 105: Fitting a Covariance Selection Model to a Matrix , 1977 .

[3]  Marco van Akkeren,et al.  A GARCH forecasting model to predict day-ahead electricity prices , 2005, IEEE Transactions on Power Systems.

[4]  A. Papalexopoulos,et al.  Forecasting power market clearing price and quantity using a neural network method , 2000, 2000 Power Engineering Society Summer Meeting (Cat. No.00CH37134).

[5]  T. Speed,et al.  Gaussian Markov Distributions over Finite Graphs , 1986 .

[6]  M. Rathmann,et al.  Do support systems for RES-E reduce EU-ETS-driven electricity prices? , 2007 .

[7]  Carlos E. Pedreira,et al.  Neural networks for short-term load forecasting: a review and evaluation , 2001 .

[8]  H. Akaike.  Likelihood of a model and information criteria , 1981 .

[9]  Paul Glasserman,et al.  Monte Carlo Methods in Financial Engineering , 2003 .

[10]  Geoffrey C. Fox,et al.  A deterministic annealing approach to clustering , 1990, Pattern Recognit. Lett..

[11]  C. D. Gelatt,et al.  Optimization by Simulated Annealing , 1983, Science.

[12]  Carbon Risk Management , 2006, 2006 International Conference on Probabilistic Methods Applied to Power Systems.

[13]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1992, Math. Control. Signals Syst..

[14]  A. Gjelsvik,et al.  Generation scheduling in a deregulated system. The Norwegian case , 1999 .

[15]  Wolfgang Eichhammer,et al.  Designing National Allocation Plans for Eu-Emissions Trading — A First Analysis of the Outcomes , 2004 .

[16]  H. Mori,et al.  Deterministic Annealing Clustering for ANN-Based Short-Term Load Forecasting , 2001, IEEE Power Engineering Review.

[17]  Peter E. Hart,et al.  Pattern classification and scene analysis , 1974, A Wiley-Interscience publication.

[18]  H. Mori,et al.  Application of Preconditioned RBFN to Temperature Forecasting for Short-term Load Forecasting , 2006, TENCON 2006 - 2006 IEEE Region 10 Conference.

[19]  C. Hirschhausen,et al.  First evidence of asymmetric cost pass-through of EU emissions allowances: Examining wholesale electricity prices in Germany , 2008 .

[20]  Christian Borgelt,et al.  Graphical models - methods for data analysis and mining , 2002 .