Grey Prediction Model of Interval Grey Numbers Based on a Novel Compound Function Transformation

Focusing on the prediction accuracy of interval grey numbers, the primary goal of this paper is to investigate novel grey prediction models to predict four typical kinds of interval grey numbers sequences respectively. The models used in this study can be briefly described as a combination of function transformation and Grey Model (GM) of interval grey numbers. According to different interval grey numbers sequences, the approaches can be applied to the upper bound and lower bound sequences of interval grey numbers or the kernel and measurement sequences of interval grey numbers. Finally, these new improved models have been verified by numerical examples and cases to demonstrate their validity and practicability. It proves these new models are not only applicable for increasingly high growth sequences where traditional grey models are not effective, but also control the enlargement of grey degrees of interval grey numbers which is very important for interval grey numbers' forecasting. In summary, the paper effectively extends function transformation technology to the field of interval grey numbers for grey prediction.