Find in Library

The main objective of the present study has been paid on segmentation of international petroleum product market of Iran based on panel data econometric approach and a new hybrid approach based on data mining algorithm. For this purpose, in the first step data of 35 gasoline importing countries in the Middle East, Asia and the Pacific between 1995 and 2012 were analyzed and through the seven hypotheses the effect of the identified components on gasoline imports have been examined. The results indicate that the market size, economic development, the structure of the free market, pump gasoline price and income elasticity of demand for gasoline have direct and significant impact on gasoline imports. In a second step, by using the identified construct and data mining algorithm named Classification and Regression Tree, international gasoline market were segmented and sequences of transition of importing gasoline countries to different segments identified. Then, sequences segmented by using data mining algorithm such as K-Means, Kohonen and two step. Analyzing quality of segments in terms of Silhouette index shows the superiority of Kohonen Kohonen, so chosen as a suitable segmentation method of sequence.