I want to use the double bootstrapping as suggested by Simar and Wilson methodology. How to calculate the double bootstrapping with truncated regression? How to report and explain Algorithm 1 and Algorithm 2 tables? Should the calculated values of various variables be same in both the Algorithms?