Why are total effects sometimes more than 1 in variable importance?

[#1]

Hello,

I'm analyzing a DoE and had a look on the variable importance graphs. Surprisingly, the sum of the total effects of my parameters are greater than 1 sometimes (see picture below).

I did not find anything in the documentation for this. Can someone explain, why the sum of the total effects can be greater than 1?

    

Best,

Christian

[#2]

These indices are used for assessing the relative importance. Neither the main effect indices or total effect indices are constrained to sum to 1.

See the statistical details here.