臺灣大學: 企業管理碩士專班郭佳瑋柏洛賓Bunker, RobinRobinBunker2013-04-082018-06-292013-04-082018-06-292010http://ntur.lib.ntu.edu.tw//handle/246246/256625In order to develop a regression model to forecast iron-ore price, CVRD & Baosteel annual contract iron ore prices were used for IOF price i.e., the dependent variable. Twelve factors were identified to have influence on the IOF price as independent variables in the regression model. In the process of developing Multi Regression Model, assumptions of – Linearity, Independence, Normality and Equal Variance were tested. It was found that multicollinearity among independent variables was the main problem. Stepwise regression was proposed to resolve this. The stepwise procedures successfully solved the problem of multicollinearity by reducing the total number of independent variables to four. The variables selected by stepwise regression were Oil Price, Production of Steel in China, World Steel Exports and China Iron Ore Production. The adjusted coefficient of correlation remained almost same. The results obtained with Stepwise Regression Model were very encouraging for years 2006-2007 and further research was suggested to overcome certain limitations.1919202 bytesapplication/pdfen-US鐵礦回歸模型預測鐵礦石Iron-oreMulti-Regression ModelForecastingthesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/256625/1/ntu-99-R97749059-1.pdf