关键词: 货运需求, Pearson相关性分析, 不确定理论, 多元线性回归, 预测

Abstract: Reasonable prediction of railway freight demand is the decision-making basis for railway management departments to build and operate. Railway freight demand is uncertain. In order to cope with the complex changes of railway freight demand, the seven key factors of railway freight demand are selected based on the Pearson correlation analysis, and uncertain multivariate linear regression model is constructed by using the uncertainty theory. The corresponding railway freight value is changed from the traditional single value to the possible demand range, which is more in line with the actual demand of railway freight transportation under the uncertain environment. According to the relevant data of the National Bureau of Statistics from 2004 to 2016, the empirical study is carried out, and compared with the prediction results of regression model and BP model. The results show that the prediction results of uncertain multivariate linear regression can deal with the uncertainty well and the accuracy is higher.

Key words: freight demand, pearson correlation analysis, uncertainty theory, multiple linear regression, prediction