摘要
循环子空间回归(CSR)通过改变解空间的维数,可以获取一系列的回归模型,其中包括最小二乘回归(LSR)、主成分回归(PCR)、偏最小二乘回归(RLSR)和许多中间回归,从中可挑选最优回归模型.本文将分析CSR的原理,给出一种可行的快速的CSR算法(RCSR),以提高计算速率和精度。
By changing the dimension of solution space, cyclic subspace regression (CSR) gets a serial of regression models which contains least square regression, principal component regression, partial least square regression and other medial regression outcomes. Then the optimal model can be picked out according to a certain criterion. This paper analyzes cyclic subspace regression, and gives a rapid algorithm of cyclic subspace regression (RCSR) to reduce the calculation. It has been successfully applied to the quantitative structure - activity relationship simulation of pesticides.
出处
《分析化学》
SCIE
EI
CAS
CSCD
北大核心
1999年第12期1386-1390,共5页
Chinese Journal of Analytical Chemistry
关键词
循环子空间回归
快速算法
定量构效关系
建模
Cyclic subspace regression, rapid algorithm, quantitative structure-activity relationship, modelling