Hurst Exponent and the Efficiency of the Czech Electricity Market

Efficient Market Hypothesis (EMH) assumes a random, independent movement of the investment instrument prices. EMH validity testing, especially on stock and currency markets, did not bring clear results. Part of the research conducted so far confirms the validity of EMH, part of the research refutes the validity of EMH. This paper focuses on commodity market efficiency testing, specifically on the electricity market in Czechia (Czech electricity and gas market operator OTE) using a time-varying Hurst exponent estimated by the Detrended Fluctuation Analysis in period 2016-2018 applied on the returns of transformed hourly intra-day electricity prices. Since the value of 0.5 for the Hurst exponent is associated with the random process, and thus with the weak form of EMH, its value can indicate whether the studied market is efficient or not. To conduct our research, we computed the Hurst exponents in the subsamples that slid along the full sample, which is called the sliding-window technique, and thereby we were able to get a view of temporal behavior of the Hurst exponent. First, our results showed the so-called multi-scale nature of the returns of transformed electricity prices, i.e. we detected two Hurst exponents on different time scales separated by the crossover. Secondly, the Hurst exponent temporal behavior, calculated on the smaller time scale, revealed the short periods when the studied market approached the weak-form efficiency. On the other hand, for the larger time scale, the Hurst exponents attained values associated with the mean-reverting process and not the random one, and thus in that case the short-term electricity market in Czechia was inefficient with respect to the weak form of EMH.