Abstract: In order to measure the similarity in the change trends between sequences, the grey change rate relational analysis (C-GRA) model has been established. The properties of the C-GRA model such as are studied in this paper. To research the consistency of the change direction between sequences, a new method for space decomposition has been put forth based on the C-GRA model with sign functionssgn (y₀(t_k)·y_i(t_k)). According to the consistency of positive or negative changes, the space decomposition index was proposed as P_i=sgn₂[sgn₁y₀(t_k)·y_i(t_k)], and the space was decomposed. Then, a grey trend analysis on the characteristic and behavioral sequences was performed to identify the sequences that shared the same step with the characteristics sequence. It should be noted that the entire ranking of the behavior sequences are not the only data that can be obtained;rather, the rankings in separate spaces can also be obtained, so the key factors in the promotion subspace, the inhibition subspace, and the coordinate subspace are clear. In the last part of this paper, the factors that affect the production of coal in the city of Fuxin are studied using the grey trend analysis, as based on the C-GRA model. Date is from the government of Fuxin. This application is presented to illustrate the effectiveness and practicality of the proposed model.

Key words: grey relational drgree with rate of change, space decomposition, trend analysis