A New Adaptive Conformable Fractional Order Grey Prediction Model with a Time Power Term and its Application

In this study, a new adaptive conformable fractional order grey prediction model with a time power term is constructed to further advance the field of grey prediction. In the modeling process, the discretization operation and ridge regression are considered to optimize the unbiasedness and adaptive performance of the model, respectively. Then, the particle swarm optimization is used to search for suitable hyperparameters. The research findings reveal that the proposed model possesses both uniformity and unbiasedness, highlighting its theoretical superiority. To verify the effectiveness of the proposed method, the real case is used. Experimental results show that the proposed method is superior to all benchmark algorithms, which verifies its effectiveness.