I do not think this is in any way a natural pivot table operation.
There appear to be two points at which you can attack the problem. The first is to duplicate the data and then reverse the Co and RC columns for the copy while changing the sign of the balance. The new dataset is then appended to the original. On refresh, the pivot table will show an antisymmetric table of values representing the differences.
The second approach is to form the existing PT and then extract the data using GETPIVOTDATA with array parameters rather than the default hard-wired values. This again allows the array to be subtracted from its transpose to give a table of differences.
(Edited to substitute CSE arrays for dynamic arrays)
I do not think this is in any way a natural pivot table operation.
There appear to be two points at which you can attack the problem. The first is to duplicate the data and then reverse the Co and RC columns for the copy while changing the sign of the balance. The new dataset is then appended to the original. On refresh, the pivot table will show an antisymmetric table of values representing the differences.
The second approach is to form the existing PT and then extract the data using GETPIVOTDATA with array parameters rather than the default hard-wired values. This again allows the array to be subtracted from its transpose to give a table of differences.
(Edited to substitute CSE arrays for dynamic arrays)
Thank you so much Peter.
The explanation on being this not being a natural pivot opertation has answered my doubt.
On the other two ways, will try same.
what i have done was i paste special pivot data on seperate space and built a vlookup based with Row & Column. GETPIVOTDATA may assist avoiding a pastespecial.
