Quintile multiple regression with fuzzy coefficients and initial Z-information

Author:

Poleshchuk Olga

Abstract

The goal of the paper is to develop a regression model under the initial Z-information based on an alternative method to the least squares method, and free from the assumptions regarding probability distributions of initial data. Formalization of input and output information is carried out on the basis of Z-numbers and linguistic variables, followed by the construction of a multidimensional quintile regression model with fuzzy coefficients. The optimization function is defined as the sum of the loss functions for the differences between the weighted output fuzzy numbers and the weighted model fuzzy numbers. To determine the parameters of the unknown regression coefficients, a linear programming problem is solved to find the minimum of the optimization function. The developed Z-regression is free from the shortcomings of existing models and provides new opportunities for solving tasks in problem areas with the active participation of experts, taking into account the reliability of information received from them.

Publisher

EDP Sciences

Subject

General Medicine

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3