In reviewing the recently answered, below question Multi-Period Contribution I think the top answer (calculating weights at the end of each period) results in a method which does allow for multi-period contribution which can be summed to the top line. Below is a screenshot example based on the linked question above which outlines the approach
So what is happening here is that month 1 return contribution remains unchanged, it is .8*.0125 + .2*.005. Month 2 however is different, i allow the 'policy' weight to drift from 1 to 1.011, then i apply the rebalance weights of 80% asset a and 20% asset b to get 'new' starting month 2 weights of .8088 and .2022 respectively. From there i calculate month 2 returns and things tie exactly.
However, to use this for my work i'm hoping to find at a paper which discusses the pros/cons of this approach and every time i search i get multi-period return attribution and not contribution discussions. Help would be greatly appreciated on naming just what this methodology is precisely.