论文部分内容阅读
为克服传统客运量回归预算法中测忽视历史数据对未来预测的影响,以及无法处理影响客运量预测不确定因素的问题,提出了采用加权盲数回归方法进行公路客运量预测。该方法以不同权重区分历史数据的差别,采用加权最小二乘法确定回归系数;同时用盲数的形式来表达预测模型的相关变量,来获得客运量可能出现的多个区间,即各个区间可能出现的可信度情况。算例表明,该方法的预测结果比较合理、可靠,预测可信度高。
In order to overcome the impact of neglecting historical data on the prediction of the future in the regression of the traditional passenger traffic forecasting method and the inability to deal with the uncertainties that affect the forecast of the passenger traffic volume, a weighted blindness regression method is proposed to predict the amount of passenger traffic on the highway. The method distinguishes the differences of historical data by different weights, and uses the weighted least squares method to determine the regression coefficients. At the same time, the relevant variables of the prediction model are expressed in the form of blind numbers to obtain the possible multiple sections of passenger traffic, that is, each section may appear The credibility of the situation. The example shows that the forecasting result of the method is reasonable and reliable, and the forecasting reliability is high.