Incorrect LOOKUP() Results with Formula in Lookup_Vector

ID: Q109978

The information in this article applies to:

Microsoft Excel for Windows, versions 2.01, 2.1, 2.10c, 2.10d, 3.0, 4.0, 4.0a, 5.0
Microsoft Excel for OS/2, version 2.2, 3.0
Microsoft Excel for the Macintosh, versions 2.20, 2.2a, 3.0, 4.0

SUMMARY

The Microsoft Excel lookup functions VLOOKUP(), HLOOKUP(), and LOOKUP() may return incorrect results if the lookup vector is the result of a formula.

CAUSE

The incorrect result in cell A18 of the example below is caused by a rounding error due to the use of the IEEE 754 floating-point standard. The IEEE 754 standard is a method of storing floating-point numbers in a compact way that is easy to manipulate. This standard is used by Intel coprocessors and most PC-based programs that implement floating-point math.

This is expected behavior and is not a bug or limitation of Microsoft Excel.

WORKAROUNDS

To work around this problem, do either of the following:

Select the lookup vector (A1:B16 in the example below), copy it, and choose the Paste Special command from the Edit menu, and then select the Values option.

-or-

Incorporate the ROUND() function in your formula. For example, to correct the example in the "Steps to Reproduce Problem" section of this article, enter the following formula in A2:A16:
=ROUND(A1+0.1,2)

MORE INFORMATION

Steps to Reproduce Problem

On a new worksheet, type the following:


   A1:  .1          B1:  1
   A2:  =A1+0.1     B2:  2
   A3:  =A2+0.1     B3:  3
   A4:  =A3+0.1     B4:  4
   A5:  =A4+0.1     B5:  5
   A6:  =A5+0.1     B6:  6
   A7:  =A6+0.1     B7:  7
   A8:  =A7+0.1     B8:  8
   A9:  =A8+0.1     B9:  9
   A10: =A9+0.1     B10: 10
   A11: =A10+0.1    B11: 11
   A12: =A11+0.1    B12: 12
   A13: =A12+0.1    B13: 13
   A14: =A13+0.1    B14: 14
   A15: =A14+0.1    B15: 15
   A16: =A15+0.1    B16: 16
   A17:
   A18: =VLOOKUP(0.3,A1:B16,2)

With the formula in A18, you would expect an exact match to be found in A3, which would return the value 3 in B3; however, VLOOKUP() returns the value 2 in B2.

For more information regarding the IEEE 754 floating point standard and rounding errors, query in this knowledge base on the following keywords:

floating-point and standard and rounding and IEEE 754

Additional query words: 4.00a

Keywords :
Version : MACINTOSH:2.20,2.2a,3.0,4.0; WINDOWS:2.01,2.1,2.10c,2.10d,3.0,4.0,4.0a,5.0
Platform : MACINTOSH WINDOWS
Issue type :