首页> 外国专利> METHOD FOR EXTRACTING MULTIPLE REGRESSION EQUATION

METHOD FOR EXTRACTING MULTIPLE REGRESSION EQUATION

机译:提取多元回归方程的方法

摘要

PROBLEM TO BE SOLVED: To provide a method for efficiently evaluating data obtained using a multivariate analysis technique and extracting an accurate multiple regression equation with a small man-hour.;SOLUTION: External data comprising a plurality of multiple regression equations obtained by performing multivariate analysis on a data group which are objective variables and explanatory variables of the objective variables in the multivariate analysis, and its evaluation data is acquired (step S1). Then, for each of the plurality of multiple regression equations, based on a predetermined multiple regression equation evaluation criterion, first evaluation which determines the adaptability of the multiple regression equations is performed (step S2). For each of the explanatory variables used for deriving the multiple regression equations which are determined to be adaptable in the first evaluation based on a predetermined explanatory variable evaluation criterion, second evaluation which determines adaptability as explanatory variables is performed (step S3). A multiple regression equation having the highest evaluation value in the second evaluation is extracted (step S4).;COPYRIGHT: (C)2010,JPO&INPIT
机译:解决的问题:提供一种有效地评估使用多元分析技术获得的数据并以较少的工时提取准确的多元回归方程的方法。解决方案:外部数据包括通过执行多元分析而获得的多个多元回归方程在作为多元分析中的目标变量和目标变量的解释变量的数据组上,获取其评估数据(步骤S1)。然后,对于多个多元回归方程式中的每一个,基于预定的多元回归方程式评估标准,执行确定多个回归方程式的适应性的第一评估(步骤S2)。对于用于基于预定的解释变量评估标准来推导确定为适合于第一评估的多个回归方程的每个解释变量,执行将适应性确定为解释变量的第二评估(步骤S3)。提取第二次评估中评估值最高的多元回归方程(步骤S4)。;版权:(C)2010,JPO&INPIT

著录项

  • 公开/公告号JP2010128779A

    专利类型

  • 公开/公告日2010-06-10

    原文格式PDF

  • 申请/专利权人 KANSAI ELECTRIC POWER CO INC:THE;

    申请/专利号JP20080302626

  • 发明设计人 IKEDA TOSHIO;

    申请日2008-11-27

  • 分类号G06F17/18;G06Q50/00;G06Q90/00;

  • 国家 JP

  • 入库时间 2022-08-21 19:03:06

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号