Hello everyone! Could someone please help?

I have a spreadsheet containing rows of transactions.
Some of the transactions are reversals. Is there a method to net out/sum out/zero out the original transaction with the reversal (that matches to the transaction number) such that I do not have to perform this task manually.
I've copied a portion of the spreadsheet below.

Fiscal Month Trans Number Cost
2016-M01 0306016265 9834.91
2016-M01 0306029857 8429.93
2016-M01 0305601550 (1292.59)
2016-M02 0306052419 2060.56
2016-M02 0306052419 2585.18
2016-M02 0305954989 1381.45
2016-M02 0305954989 1001.05
2016-M02 0305954989 (1381.45)
2016-M02 0305954989 (1001.05)
2016-M02 0306078110 3877.77
2016-M02 0305954989 1381.45
2016-M02 0305954989 (1381.45)
2016-M02 0305954989 1001.05
2016-M02 0305954989 (1001.05)
2016-M02 0306016265 (9834.91)
2016-M02 0306052419 (2060.56)
2016-M02 0306052419 (2585.18)
2016-M02 0305954989 (1381.45)
2016-M02 0305954989 1381.45
2016-M02 0305954989 (1001.05)
2016-M02 0305954989 1001.05
2016-M02 0306078110 (3877.77)
2016-M02 0306085310 1053.74
2016-M03 0304356338 2985.60
2016-M03 0304356338 (2985.60)
2016-M03 0306093769 1292.59
2016-M03 0306118371 35000.00
2016-M03 0306134887 7376.19
2016-M03 0306134887 2650.37
2016-M04 0305974344 (1053.74)
2016-M04 0305746045 (8429.93)
2016-M04 0306118371 (35000.00)
2016-M04 0306134887 1472.84
2016-M04 0306134887 (1472.84)
2016-M04 0306134887 (7376.19)
2016-M04 0306134887 (2650.37)

I currently use this formula COUNTIFS(E$2:$E3,E3)<=MIN(COUNTIFS($E$2:$E$9998,-E3),(COUNTIFS($E$2:$E$9998,E3))))

It is useful to match both corresponding positive and negative value, but it doesn't match to the corresponding transaction number. How do I add additional condition so that I can find out the 'true' reversal transaction?

Thank you guys!