I have a Pivot table that has a variable (set to Count) that has columns over a 10 year period. The last year/period represents only a potion of the year so I would like to "annualize" the last year/period data. I want to do this by dividing the last years count value by the ratio of days currently represented in the year to date (e.g., .141 for date in February). I've looked at calculated variables but it does not appear that you can work with a count value and modify it as I've described. I reuse the same pivot table for different data sets that (using VBA) that prohibits me from putting the pivot table value in a normal Excel cell and doing the calculations. I pull the values directly from the pivot table to a data select on an Excel chart.

I will look at how I might display the PT. The underlying data is really not an issue and should have no impact on the solution. I simply have a variable in the PT that I've set to Count. The PT summarizes the number of rows associated with that variable as a number (count). I have a count for each of the 10 years of data I'm analyzing. Since I don't have a full year for 2011 I want to correct the 2011 count by dividing by a correction factor. As I noted, I cannot pull the 2011 count to a cell and do the calculations as the overall automated analysis procedure requires I pull the data directly to an Excel chart process. The exact PT path for each chart value is inserted into the "select data" function of the Excel chart. I need to find a way to correct one of the values before plotting. I've tried to add calculations into the select data area but it isn't compatible with this type function. I've also looked at using a calculated value but don't see a way to use a count (of a variable) in the calculation.

I understand, but it would be a simple think for you to create a simple problem I can help you to solve, as opposed to me straining my brain to simulate that. I'm willing to help you solve the problem - are you willing to create a sample dataset so I can help you solve the problem?

Dave

0

Can Concerto Cloud Services help you focus on evolving your application offerings, while delivering the best cloud experience to your customers? From DevOps to revenue models and customer support, the answer is yes!

Ok - I took a bit of time and thought about the problem. Then, I created my own sample dataset - so I hope this suits. Sometimes by creating a sample (data does NOT matter, that's correct - but pictures in the mind saves thousands of words - or the 339 you used to describe the problem :)

Anyway, I hope this works for you. I created a simple dataset with two people, and salaries over a 5 year period (how much was earned between two dates - start of year to end of year. In the final year, the ending date was earlier than December 31.

I then created a couple columns in the data (for check purposes only, as I realize you cannot create these in your dataset!) One for the fraction of the year, then dividing by the salary earned achieved the ANNUALIZED salary.

Ok, once satisfied I had something to solve for, I created the pivot table. Then, I created 2 calculated fields (I assume you are familiar with these as your question uses this language, so I'll forgo what buttons to press to get there).

YearFraction = (end - start +1)/365 - not counting leap year as this is an example
AnnualizedSalary = Salary / YearFraction

Then, I put these in the pivot table and checked against the table I created in the spreadsheet and (presto) they were the same.

Here's hoping this is your solution :)

Please see attached demo example. Please feel free to ask questions, if needed.

If this is the not direction you're going, or you want something more precisely like what you're trying to do, then please provide SAMPLE DATA to help with clarification.

Thanks for your indulgence as it will help speed responses and solutions!

If the solution doesn't suffice and you need help generating sample data or example - you could provide a print screen of your pivot table or even double click one of the value fields to generate a simple 1-row table that could be useful in assisting you further.

Cheers,

Dave

0

jaustinrAuthor Commented:

Dave,
I guess I should have used all 500 words (or a lot less) to describe my problem. :) I really appreciate your taking the time to look at my issue. You are exactly on the path of what I'm trying to do. In your example, I have the year fraction. Since I do an update 1-2x/month, so this is not an issue. However, my "salary" variable is actually a "count" of salaries. So, my real problem is how to do additional calculations (within PT) on the count. At this point, my only solution may be to create the annualized data point back in Access where my original dataset resides. However, not sure how that would interact with my PT and it makes the process a lot more complex.
Thanks again and any additional thoughts appreciated.

Ok - willing to try this once more, without data - which would be really refreshing. Many times we can give you exactly what you asked for, and yet that's not what you're LOOKING for. A PICTURE is worth 1000 words and sometimes 1000 words don't get it :)

Here it goes. I assume there's two people, with fraction thru the year called Annualized Fraction (which you say you have) and have received # of paychecks (no ryme or reason, that's just how many times they got paid).

So the # paychecks is your COUNT of salaries, for example purposes.

If I'm off here, I'd like you to at least show an example to me just like how I'm explaining to you.

So if Mary has had 12 paychecks and her annualizing fraction = .5, the Annualized # Paychecks would be 24.

Here's the example, see attached range 36-48 of the data and pivot table. I clicked in the data, then selected Pivot Table tools, then Formulas, and added a Calculated Field, called Annualized Count, with the formula above, see picture. Also, see attached.

Just returned from travel and had a couple of things get in the way. Plan on trying today or tomorrow. Will let you know my results.

Thanks,

John

0

jaustinrAuthor Commented:

Dave,

Sorry for the delay. The problem with your solution is that you have a count variable in your original dataset. My count comes from using a variable in the data and using the Value Field Settings window to set the type calculation to Count (instead of sum, max min, etc.). So when I use your approach to go to Tools/Formulas/Calculated Field, the only variable that show are the original variables, not the count of a variable (or any of the other standard variable calculations). I've played around using a "count(variable name)/annualize factor" in the formula field (for calculated value). This seemed to give me a proper value, but it creates additional problems. I have grouped years and it appears that the calculated value makes some changes in the PT cache.

Thanks for thinking about this and providing recommendations.

Dave