CMStatistics 2016: Start Registration
View Submission - CMStatistics
B0919
Title: Variable selection to predict electricity price categories Authors:  Camino Gonzalez - Universidad Politécnica de Madrid (Spain) (Spain) [presenting]
Jose Mira - Universidad Politecnica de Madrid (Spain)
Julia Garcia Lezana - Viesgo (Spain)
Abstract: A legal framework has been recently implemented in the Spanish electricity market under which residential consumers pay according to the hourly price of electricity. For consumers it could be more useful to know, rather than the exact value of the hourly electricity price, the interval or category where this price belongs, for instance high, medium and low, in such a way that the consumers can also implement consumption shift strategies depending on the price signals received at home. Both design of experiments and data mining techniques have been applied to identify the variables with the highest influence on threshold values of price categories as well as optimal settings for these thresholds.