Average data based on unique change in date, ignoring time
 

Hi,

I have a question that is probably a combination of VBA & Excel formulas, so if I put this in the wrong location, please let me know & I'll move it.

Anyway, I have a CSV data file with 3 columns:

Column 1 Column 2 Column 3
(Text) Date (in the following format) Data (floating point)
a 9/19/2013 12:02:00 AM 6.58667
b 9/19/2013 12:03:00 AM 6.44361
c 9/19/2013 12:04:00 AM 6.32134
d 9/19/2013 12:05:00 AM 6.54321
e 9/20/2013 7:26:00 PM 10.11111
f 9/20/2013 7:27:00 PM 10.23345
g 9/20/2013 7:28:00 PM 10.54321
h 9/20/2013 7:29:00 PM 10.67890
i 9/21/2013 11:58:00 PM 3.12345
j 9/21/2013 11:59:00 AM 3.54321
k 9/21/2013 12:00:00 PM 4.11111
l 9/21/2013 12:01:00 PM 4.54321


The goal is to take an average of the Data in Column 3 for each instance when the Date in Column 2 changes, ignoring the time.

So, in the example above, the result would be 2 more columns, each with the following data:

Column 4 Column 5
(Unique Dates) (Average of each Date)
9/19/2013 6.4737075
9/20/2013 10.3916675
9/21/2013 3.830245

If this can be automated such that, when the CSV file is pulled into Excel, a macro can be run to produce Columns 4 & 5, that would be the most ideal solution. I am confused as to what commands in VBA would use only the unique instances of the date, ignoring the time.

If not, any assistance would be appreciated.

Thanks in advance,
Rob
Excel 2007

P.S. sorry that the formatting isn't coming out correctly - please let me know what I'm doing wrong & I'll correct it.

Average data based on unique change in date, ignoring time
 

Hi,

Am Tue, 8 Oct 2013 17:37:25 +0100 schrieb rbrookov:

Column 1 Column 2 Column 3
(Text) Date (in the following format)
Data (floating point)
a 9/19/2013 12:02:00 AM 6.58667
b 9/19/2013 12:03:00 AM 6.44361
c 9/19/2013 12:04:00 AM 6.32134
d 9/19/2013 12:05:00 AM 6.54321
e 9/20/2013 7:26:00 PM 10.11111
f 9/20/2013 7:27:00 PM 10.23345
g 9/20/2013 7:28:00 PM 10.54321
h 9/20/2013 7:29:00 PM 10.67890
i 9/21/2013 11:58:00 PM 3.12345
j 9/21/2013 11:59:00 AM 3.54321
k 9/21/2013 12:00:00 PM 4.11111
l 9/21/2013 12:01:00 PM 4.54321

The goal is to take an average of the Data in Column 3 for each instance
when the Date in Column 2 changes, ignoring the time.

in D1 try:
=IF(COUNTIF($B$1:$B1,B1)=1,AVERAGE(OFFSET(B1,,2,CO UNTIF($B$2:$B$12,B2))),"")
and copy down


Regards
Claus B.
--
Win XP PRof SP2 / Vista Ultimate SP2
Office 2003 SP2 /2007 Ultimate SP2

Average data based on unique change in date, ignoring time
 

rbrookov wrote:

Hi,

I have a question that is probably a combination of VBA & Excel
formulas, so if I put this in the wrong location, please let me know &
I'll move it.

Anyway, I have a CSV data file with 3 columns:

Column 1 Column 2 Column 3
(Text) Date (in the following format)
Data (floating point)
a 9/19/2013 12:02:00 AM 6.58667
b 9/19/2013 12:03:00 AM 6.44361
c 9/19/2013 12:04:00 AM 6.32134
d 9/19/2013 12:05:00 AM 6.54321
e 9/20/2013 7:26:00 PM 10.11111
f 9/20/2013 7:27:00 PM 10.23345
g 9/20/2013 7:28:00 PM 10.54321
h 9/20/2013 7:29:00 PM 10.67890
i 9/21/2013 11:58:00 PM 3.12345
j 9/21/2013 11:59:00 AM 3.54321
k 9/21/2013 12:00:00 PM 4.11111
l 9/21/2013 12:01:00 PM 4.54321


The goal is to take an average of the Data in Column 3 for each instance
when the Date in Column 2 changes, ignoring the time.

So, in the example above, the result would be 2 more columns, each with
the following data:

Column 4 Column 5
(Unique Dates) (Average of each Date)
9/19/2013 6.4737075
9/20/2013 10.3916675
9/21/2013 3.830245

If this can be automated such that, when the CSV file is pulled into
Excel, a macro can be run to produce Columns 4 & 5, that would be the
most ideal solution.

If not, any assistance would be appreciated.

Thanks in advance,
Rob

P.S. sorry that the formatting isn't coming out correctly - please let
me know what I'm doing wrong & I'll correct it.





regular pivot table where dates go to rows and column3 as value with
average as agregating function ?

Average data based on unique change in date, ignoring time
 

Hi,

Am Tue, 8 Oct 2013 20:17:38 +0200 schrieb Claus Busch:

in D1 try:
=IF(COUNTIF($B$1:$B1,B1)=1,AVERAGE(OFFSET(B1,,2,CO UNTIF($B$2:$B$12,B2))),"")

sorry, there is a typo in the formula above.
Try:
=IF(COUNTIF($B$1:$B1,B1)=1,AVERAGE(OFFSET(B1,,2,CO UNTIF($B$1:$B$12,B2))),"")

Or if you prefer to have the result in the row of the last occurence try
in D1:
=IF(B2<B1,AVERAGE(OFFSET(B1,,2,-COUNTIF($B$1:B1,B1))),"")
and copy down.


Regards
Claus B.
--
Win XP PRof SP2 / Vista Ultimate SP2
Office 2003 SP2 /2007 Ultimate SP2

Claus,

Hi, thank you for responding. Yes, that is a technique which works quite fine in Excel. However, as this task is to be done over & over again, and on many hundreds, perhaps thousands of rows, each time it is done, I would think that something a bit more automated might be more efficient.

However, I appreciate you providing an excellent, and correct way of performing the Average just by using Excel's built-in commands.

Thanks!

Witek,

Yes, a pivot table is a way to do it. However, is there a way to do it utilizing a macro, and without creating a pivot table? I know that a macro can be written which produces a pivot table, but can a macro be written such that the result is not in a pivot table? Or is that not possible?

Thank you for your suggestion!!
Rob

Average data based on unique change in date, ignoring time
 

Hi,

Am Tue, 8 Oct 2013 20:06:03 +0100 schrieb rbrookov:

Hi, thank you for responding. Yes, that is a technique which works
quite fine in Excel. However, as this task is to be done over & over
again, and on many hundreds, perhaps thousands of rows, each time it is
done, I would think that something a bit more automated might be more
efficient.

then name your data with a dynamic range name and create a pivot table
from this name.


Regards
Claus B.
--
Win XP PRof SP2 / Vista Ultimate SP2
Office 2003 SP2 /2007 Ultimate SP2