And to apply that information to this specific instance, the binary
approximations involved are
90413.630000000004656612873077392578125
-90150.979999999995925463736057281494140625
- 262.6499999999999772626324556767940521240234375
--------------------------------------------------
0.0000000000087538865045644342899322509765625
which Excel correctly displays to its documented limit of 15 figures as
0.00000000000875388650456443

Bottom line: the math is correct, but most decimal fractions have no
exact binary representation and hence must be approximated. With
decimal fractions, digits beyond the 15th may not be what you expect and
subtraction may reveal evidence of those approximations. In cases like
this, where you know that anything beyond a certain decimal place is
residue from binary approximations to decimal fractions, then you should
round results to that number of places to avoid surprises.

Jerry

Debra Dalgleish wrote:

The information in the following MSKB article may help explain the problem:

Floating-point arithmetic may give inaccurate results in Excel
http://support.microsoft.com/kb/78113

SundanceKidLudwig wrote:

Debra,
Thank you for getting back to me on this. The thing is the formulas
involved are only addition and subtraction at 2 decimal places. I
rechecked each cell involved by adding or subtracting the cell amount
to ensure it was zero and all cells except 1 showed "-" as the total.
The cell in question is a formula that takes the result of 1 cell,
90,413.63 and subtracts 90,150.98 which equals 262.65. If I subtract
262.65 from formula in that cell I get 0.00. When I expand it to 24
places the resulting number is 0.000000000008753886504564. I
understand why the cell shows 0.00 but I don't understand how the
result of 262.65 is really 262.650000000009000000000000 especially
when the two components are only 2 decimal places. Other similar
formulas are not presenting this same problem. I don't get it. Thank
you again for any insight to this issue.