Forecasting of Electricity Prices with Hidden Markov Models and Wavelet Transform

In the competitive electricity markets, accurate price forecasting embody crucial information for producers and consumers when planning bidding strategies in order to maximize their benefits and utilities, respectively. This paper presents a brief overview of different techniques and models in the literature of energy price forecasting. Based on this approach, a new complementary classification of price forecasting models is proposed, which can be used for choosing the suitable method for certain cases. The two major groups of the models are the stationary and non-stationary time series ones. Using of Wavelet Transform and Hidden Markov Models (HMMs) have been recently proposed among these groups respectively. These two time series models have differently upgraded price forecasting. This paper proposes WT-based models and HMMs in detail with a modeling and simulation of the later in the Spanish electricity market.

Keywords:

Hidden Markov models , non-stationary time series , price forecasting , price models , stationary time series , Wavelet transform