Research on the forecasting method for natural gas price based on the data mining technique
-
-
Abstract
Natural gas price is a key factor which influences the operation decision and profit of natural gas enterprises.In this case, how to forecast the price accurately has naturally become a hot topic in the natural gas industry.Besides, with the rapid development of data mining techniques, how to use the data mining techniques in natural gas price forecast also attracts lots of attentions from the academia.Firstly, the existing forecast methods of natural gas price are reviewed in this paper.Secondly, by modifying the two mechanisms in traditional pattern sequence similarity search(PSS) which is based on data mining techniques, i.e., the matching mechanism in searching historical data series and the result processing mechanism, a new adjusted pattern sequence similarity search(APSS) method is proposed in this paper to forecast the natural gas price.Thirdly, to verify the validity of the proposed method, the data of US daily natural gas spot price are used to the method.The empirical results show that the proposed APSS method can forecast the natural gas price reasonably and the forecast results of the APSS method have a higher prediction accuracy by comparing with the traditional PSS method.
-
-