Matrix Multivariate Grey Models for the Interval Number Sequence

The multivariate grey model takes into account the influencing factors of system characteristics and is more widely used than the univariate grey model. In order to make two multi-variable grey models, GM (1, N) and GM (0, N) that directly predict the interval number sequence, the parameters of the two model equations are improved into second-order matrix form, and the interval number is regarded as a two-dimensional column vector. In this way, two new grey models termed as MINGM (1, N) and MINGM (0, N) are obtained. Cramer rule and the least square method are used to solve the prediction formula. Examples of different development trends are used to test the effectiveness of the proposed model. The result of prediction is compared with two typical methods of interval prediction. The results show that the proposed models have much better performance than those of models for comparison.