請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/47812
標題: | 利用多元回歸模型預測鐵礦石價格 Forecasting Iron-ore Prices Using Multi-Regression Model |
作者: | Robin Bunker 柏洛賓 |
指導教授: | 郭佳瑋(Chia-Wei Kuo) |
關鍵字: | 鐵礦,回歸模型,預測,鐵礦石, Iron-ore,Multi-Regression Model,Forecasting, |
出版年 : | 2010 |
學位: | 碩士 |
摘要: | In 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. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/47812 |
全文授權: | 有償授權 |
顯示於系所單位: | 管理學院企業管理專班(Global MBA) |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-99-1.pdf 目前未授權公開取用 | 1.87 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。