If we begin reinsurance by grouping losses to location level before the first reinsurance iteration, we can avoid allocating back to items, and use allocrule 0 for Risk Level = location iterations. This should improve performance by avoiding back-allocation to items in subsequent iterations.
If we begin reinsurance by grouping losses to location level before the first reinsurance iteration, we can avoid allocating back to items, and use allocrule 0 for Risk Level = location iterations. This should improve performance by avoiding back-allocation to items in subsequent iterations.