Multidisciplinary decision-making approach to high-dimensional event history analysis through variable reduction methods

Abstract

As an analytical approach, decision-making is the process of finding the best option from all feasible alternatives. The application of decisionmaking process in economics, management, psychology, mathematics, statistics and engineering is obvious and this process is an important part of all science-based professions. Proper management and utilization of valuable data could significantly increase knowledge and reduce cost by preventive actions, whereas erroneous and misinterpreted data could lead to poor inference and decision-making. This paper presents a class of practical methods to analyze high-dimensional event history data to reduce redundant information and facilitate practical interpretation through variable inefficiency recognition. In addition, numerical experiments and simulations are developed to investigate the performance and validation of the proposed methods.