主成分回归与分位数回归在两类数据中的应用分析-application of principal component regression and quantile regression in two kinds of data.docxVIP

哈尔 哈尔滨工业大学理学硕士学位论文 哈尔 哈尔滨工业大学理学硕士学位论文 - - PAGE IV - - - III - Abstract In the process of analysing the actual data, the multivariate statistical analysis method is widely used as the actual problem is often multivariate. The two kinds of data were researched and analyzed by the principal component regression and quantile regression in this thesis. Firstly, the characteristics of the two methods was compared, then the two methods were combined in order to analyze the data more comprehensively. The independent variables of the two sets of data this thesis selects have different levels of multicollinearity, principal component regression is a more commonly used regression method which is to overcome the multicollinearity among the independent variables. Quantile regression can get different regression equations due to the difference of the quantile. By comparing the significance and the magnitude of parameter of the quantile regression equation, estimate the effect of independent variable on the dependent variable among different quantiles. Compared with the principal component regression, quantile regression analyzes the data more comprehensively. Median component regressions to the two kinds of data selected and contrastive analyze its result and the result of principal component regression, it could be find that the collinearity between models will let the median regression equation deviate from the actual situation, which did not have reference value. Considering that the principal component in the principal component regression is linearly independent, the quantile regression of principal component is studied in this thesis. For using quantile regression to principal component can not only overcome the influence of multi-collinearity among the independent variables, but also reduce the number of variables and simplify the regression equation. Quantile regression under principal component can get each principal component of different subsites’ influence on the dep