Application of partial least squares regression in data analysis of mining subsidence
(1.Xuzhou Normal University, Xuzhou 221116, China2.Shandong University of Science and Technology, Tai’an 271019, China3.Shandong University of Technology, Zibo 255049, China)
Abstract: Based on the surveying data of strata-moving angle and the ordinary least squares regression, this paper is to construct, a regression model is constructed which is strata-moving parameter β concerning the coal bed obliquity, coal thickness, mining depth, etc. But the regression is unsuccessful. The result is that none of the parameters is suited, this is not up to objective reality. This paper presents a novel method, partial least squares regression (PLS regression), to construct the statistic model of strata-moving parameter β. The experiment shows that the forecasting model is reasonable.
Key words: strata-moving parameter； least squares regression； multi-collinear； PLS regression