Steps of Grey Relational analysis method

https://wenku.baidu.com/view/dc356290af1ffc4fff47ac0d.html?rec_flag=default&sxts=1538121950212

1. Determine factors, determine the impact value of factors, and then dimensionless treatment. (This is basically a necessary step for this type of model)

Xi=[xi1, Xi2, Xi3, ..., Xin]

2. Determine the reference series. (that is, who do you want to associate with)

Xs=[x1, x2, x3,..., xn]

3. The right to decide. (This step is not necessary, it needs to be weighted when the impact factor is biased, and if each factor is considered equal, it is no longer weighted)

4. The next step is to calculate the correlation coefficient, more troublesome

In fact, I also look dizzy ... Skip it.

5. Calculate the correlation degree.

In fact, the last step is to calculate the degree to which an object is associated with a standard under a certain factor. Then you can get the correlation ri of each object and the standard by using the weighted average of the previous weights.

6. Then sort, you can get the correlation degree order, find the most relevant object.

---------------------This article from Zhangyue_lala csdn blog, full-text address please click: 70157514?utm_source=copy

Steps of Grey Relational analysis method