问题
I am using index/match to extract data by matching date/times using INDEX(D:D,MATCH(A2,B:B,0))
A:A column of date/times has been manually entered.
B:B column of date/times input as B2+time(,30,) and copied down.
After 3 lines the formula gives #N/A as an answer.
It has something to do with the times not matching exactly even though A5=B5 gives a true result that row gives #N/A
If I copy A:A to B:B i.e. all manually entered times it works perfectly.
Please help.
A B C D E F G H I J K
AET Entered AET + 30 = volts lo i1 i0 vl1 vl0 Value AET Value Time +30
1/08/15 10:00 1/08/15 10:00 TRUE 36 36 36 36 36 36 42217.4166666667000 42217.4166666667000
1/08/15 10:30 1/08/15 10:30 TRUE 23 23 23 23 23 23 42217.4375000000000 42217.4375000000000
1/08/15 11:00 1/08/15 11:00 TRUE 44 44 44 44 44 44 42217.4583333333000 42217.4583333333000
1/08/15 11:30 1/08/15 11:30 TRUE 55 44 44 #N/A 44 #N/A 42217.4791666667000 42217.4791666667000
1/08/15 12:00 1/08/15 12:00 TRUE 13 55 55 #N/A 55 #N/A 42217.5000000000000 42217.5000000000000
1/08/15 12:30 1/08/15 12:30 TRUE 32 13 13 #N/A 13 #N/A 42217.5208333333000 42217.5208333333000
1/08/15 13:00 1/08/15 13:00 TRUE 56 32 32 #N/A 32 #N/A 42217.5416666667000 42217.5416666667000
1/08/15 13:30 1/08/15 13:30 TRUE 70 56 56 #N/A 56 #N/A 42217.5625000000000 42217.5625000000000
1/08/15 14:00 1/08/15 14:00 FALSE 43 70 70 #N/A 70 #N/A 42217.5833333333000 42217.5833333334000
1/08/15 14:30 1/08/15 14:30 TRUE 31 43 43 #N/A 43 #N/A 42217.6041666667000 42217.6041666667000
1/08/15 15:00 1/08/15 15:00 TRUE 21 31 31 #N/A 31 #N/A 42217.6250000000000 42217.6250000000000
1/08/15 15:30 1/08/15 15:30 FALSE 11 21 21 #N/A 21 #N/A 42217.6458333333000 42217.6458333334000
1/08/15 16:00 1/08/15 16:00 TRUE 66 11 11 #N/A 11 #N/A 42217.6666666667000 42217.6666666667000
1/08/15 16:30 1/08/15 16:30 TRUE 45 66 66 #N/A 66 #N/A 42217.6875000000000 42217.6875000000000
1/08/15 17:00 1/08/15 17:00 FALSE 23 45 45 #N/A 45 #N/A 42217.7083333333000 42217.7083333334000
回答1:
I'll leave @AxelRichter to post the answer in his comment, which involves rounding the timestamps and referencing the rounded numbers to avoid floating point precision.
An alternative solution would be to simply allow VLOOKUP/MATCH to pick up the 'nearest value' result, using the final argument in each function. ie:
=INDEX(D:D,MATCH(A2,B:B,1))
Keep in mind that this will only work if your data is sorted. It has the added benefit that if a missing result should naturally be the next closest row, that row is picked up - and likewise it has the added cost that it will not warn you when a match isnt found.
回答2:
This is an issue with floating point precision. As you see even the = comparison fails with 1/8/15 14:00 42012,5833333333 vs. 42012,5833333334. The exact matching with MATCH(..,0) or VLOOKUP(.., FALSE) uses an other comparison method and so it fails more often. If we need exact matching date time values then we need those values ROUNDed the same count of decimals. So we need helper columns which ROUNDs A and B to 8 decimals and then use those helper columns in MATCH(..,0) or VLOOKUP(.., FALSE).
At least one helper column is needed since we could round the lookup value directly within the functions. But the lookup array we can not round that way.
Example:
Formulas:
B3downwards:
=B2+TIME(0,30,0)
C2downwards:
=A2=B2
D2downwards:
=MATCH(B2,A:A,0)
E2 downwards:
=MATCH(ROUND(B2,8),G:G,0)
G2downwards:
=ROUND(A2,8)
来源：https://stackoverflow.com/questions/34603682/index-match-time-match-errors