摘要
[目的]探讨偏最小二乘回归的理论及其应用。[方法]应用医学实例计算,对偏最小二乘回归与主成分回归及一般最小二乘回归进行比较。[结果]偏最小二乘回归对数据的拟合度优于主成分回归和一般最小二乘回归法。[结论]偏最小二乘回归适用于处理有多重共线性的资料,当解释变量个数多、样本量少时尤为有效,是稳健的数据“软”建模的统计方法。
Objective To study the theory and application of partial least squares regression (PLS). To compare PLS with principal component regression (PCR) and ordinal least square regression (OLS) by a medical case. Result PLS is superior to PCR and OLS in goodness-of-fit for the case. Conclusion PLS is a moderate 'soft' statistical modeling method, it is fit for data with multicollinearity, and it is more effective especially when the number of independent variable is large and the sample size is small.
出处
《海峡预防医学杂志》
CAS
2005年第3期3-6,共4页
Strait Journal of Preventive Medicine
关键词
偏最小二乘回归
一般最小二乘回归
主成分回归
Partial Least Square Regression (PLS)
Principal Component Regression (PCR)
Ordinal Least Square Regression (OLS)
