Study on unbiased interval grey number prediction model with new information priority. Issue 1 (16th October 2019)

• Record Type: Journal Article
• Authors: Li, Ye; Li, Juan
• Is Part Of: Grey systems. Volume 10:Issue 1(2020)
• Page Start: 1
• Page End: 11
• Publication Date: 2019-10-16
• Subjects: Interval grey number -- New information priority -- Recursive iteration solution -- Unbiased prediction; Cybernetics -- Periodicals; Systems engineering -- Periodicals; 003.5
• DOI: 10.1108/GS-06-2019-0018
• Languages: English
• ISSNs: 2043-9377

Abstract

Abstract : Purpose: The purpose of this paper is to construct an unbiased interval grey number prediction model with new information priority for dealing with the jumping errors from difference equation to the differential equation in the prediction model of interval grey number. Design/methodology/approach: First, this study obtains a set of linear equations about the model parameters by taking the minimum error sum of squares between the accumulative sequence and its simulation values as criterion, and solves them on the basis of the Crammer rule. Then, according to the new information priority principle, it selects the last number of the accumulated generation sequence as the initial value and gives the expression of the time response function by the recursive iteration method to establish the interval grey number prediction model. Findings: This paper provides an unbiased interval grey number prediction model with new information priority, and the example analysis shows that the method proposed in this paper has higher prediction precision and practicality. Research limitations/implications: If there is a better method to whiten the interval grey number, so as to fully tap the grey information contained in it, the accuracy of the model will be higher. Practical implications: The model proposed in this paper can avoid the error caused by jumping from difference equation to differential equation and make full use of new information. It can be better used in a problem where new information has a great influence on prediction results. Originality/value: This paper selects the last number of the accumulated generation sequence as the initial value and gives the expression of the time response function by the recursive iteration method. Then, it constructs an unbiased interval grey number prediction model with new information priority.