View Single Post
 
Old 12-18-2013, 10:50 PM
macropod's Avatar
macropod macropod is offline Windows 7 32bit Office 2010 32bit
Administrator
 
Join Date: Dec 2010
Location: Canberra, Australia
Posts: 22,467
macropod has a reputation beyond reputemacropod has a reputation beyond reputemacropod has a reputation beyond reputemacropod has a reputation beyond reputemacropod has a reputation beyond reputemacropod has a reputation beyond reputemacropod has a reputation beyond reputemacropod has a reputation beyond reputemacropod has a reputation beyond reputemacropod has a reputation beyond reputemacropod has a reputation beyond repute
Default

Quote:
Originally Posted by NewPuppy View Post
Does exist any excel's addon (or any kind of software that can use Excel file extensions) which could analyze every single filter's (every single piece of chart's axis) parameter with every single cell (all columns/lines) within the same Excel file? So analyzed would be every cell with every other cell, including any and all (one by one) filters?
...
I would like get reports of uncommon variations/deviations seen on charts so I could get answer on the question similar to "Why is particular line in graph (or particular datas in tables) different/unusual than others?" or answer on the question similar to "Because of parameters of which filters and which conditions are some values in tables so much different than other ones?"
There are many statistical analysis packages available for Excel. Before you can even choose one, though, you need to know what sort of an analysis you need to run and how to interpret the results. Then it's pretty much just a matter of choosing whichever package provides the best tools within your budget. There is no point in simply throwing a set of statistical analyses at data for which the methods are inappropriate or for which you don't have all of the required data. And, of course, you need to specify the analysis inputs, parameters and filtering criteria for each and every run.

Depending on the package you choose, you may be able to automate that for different data sets, but you will have to specify all of those elements as part of the automation process. in other words, the only time saving might be in the ability to run each statistical analysis back-to-back, rather than having to wait for you to input the details for each one. There can be advantages to either approach. If you do them iteratively, you can look at the results from each pass and modify the way you do the next analysis so that useless analyses are avoided. Ultimately, though, all any stats package can do is analyse the data; none can determine the results' relevance to you.
__________________
Cheers,
Paul Edstein
[Fmr MS MVP - Word]
Reply With Quote