多元逐步回归模型在滑坡变形预测分析中的应用

在滑坡体变形监测数据分析中,引入时效、温度和降雨等作为滑坡体变形的影响因子,应用SPSS软件,采用多元逐步回归算法,建立滑坡变形的统计回归模型。将该模型拟合值与实测值的对比分析,能较好地反映滑坡体的变形规律和发展趋势,并识别影响滑坡变形的主要因素。工程实例表明,该模型的预测精度高于以往其他方法构建的模型,且操作简便。

Application of Multivariant Stepwise Regression Model in Prediction and Analysis of Landslide Deformation

In the analysis of the monitoring data of landslide deformation,a statistical regression model is established in SPSS software,in which the time effect,temperature and rainfall factors are included and the multivariant stepwise regression algorithm is adopted.Comparison of the fitted values from the regression model and the actual measurement data shows that the model can accurately reflect the deformation characteristics and the development trend,and identify the major factors affecting the landslide deformation.Practical example shows that the prediction accuracy of the model is higher than that of other methods,and the operation is simple.