Data Analysis

J

john.e.palmer

What happened to Tools/Data Analysis in Excel 2008? Things like
correlation, histogram, moving average, etc.
 
E

Eunice Goldberg

Where is histogram? Also I cannot install DDXL from Data Desk because it is no longer in the Ad-ins.
Eunice
 
B

Bob Greenblatt

Where is histogram? Also I cannot install DDXL from Data Desk because it is no
longer in the Ad-ins.
Eunice
Click to expand...
If you are talking about Excel 2008. There are no addins because there is no
VBA.
 
M

mauricev

Are you trying to tell us that the histogram functionality has been removed from the program?

Why would there be an option to manage Add-ins if there are any?
 
C

Carl Witthoft

JE McGimpsey said:
Not the functionality, just the Wizard, which required VBA. The Wizard
just used built-in functions to produce the report.

You can use the Histogram workbook here

http://www.coventry.ac.uk/ec/~nhunt/oatbran/

to get a good replacement.
Click to expand...

I have a jazzed-up version of oatbran's histogram tool with a few more
features and (IMHO :) ) a better graph. grab it at

http://home.comcast.net/~cgwcgw/excelstuff.html

Personally, I find the hist tool in R to be much better :) .
 
L

Laura Breshears

How frustrating! I upgrade to Office 2008 and lose functionality? I use TTESTs and histograms all the time in my work. Now, I will have to find software that will do them. Will Microsoft make some sort of add-in for this if we contact them enough? The oatbran thing is ok, but it cannot put two data sets in one histogram for comparisons.
 
M

Mike Middleton

Laura Breshears -

Regarding "TTEST": The wizards of the Analysis ToolPak are not in Excel
2008. The underlying worksheet function, named
TTEST(array1,array2,tails,type), is available.

Regarding "histograms": In addition to oatbran, you could create the
frequency distribution using the COUNTIF worksheet function or the
array-entered FREQUENCY function. Then create a Column chart.

Regarding "two data sets in one histogram": As I recall, the Histogram tool
in Excel 2004 works with only one data set, not two.

- Mike Middleton
http://www.DecisionToolworks.com
Decision Analysis Add-ins for Excel
 
L

Laura Breshears

Yes, I figured out the TTEST thing. Thank goodness it is still there. Office 2004 was perfectly able to put two (or more) histogram data sets into one graph so that you could directly compare the spread of data points between experimental and control data. It was very useful. I also just discovered (to my horror) that I can no longer add custom error bars to data points. So, excel 2008 is officially useless to the scientific community.
 
J

JE McGimpsey

Laura Breshears said:
Office 2004 was perfectly able to put two (or more) histogram data sets into
one graph so that you could directly compare the spread of data points
between experimental and control data. It was very useful.
Click to expand...

You can do the same thing in XL08. Just add another data series.
 
G

Gregg Larson

As I only use the data analysis functions on occasion I find that the removal of this function totally without basis. I am paid to understand my craft not how to manipulate a program such as Excel. I don't mean to vent on the MVP's as they are relaying the information regarding the program and I appreciate their efforts. Excel has now been reduced to a simple number cruncher not worth any type of detailed analysis required in engineering in my view.
 
C

Carl Witthoft

Gregg Larson said:
As I only use the data analysis functions on occasion I find that the removal
of this function totally without basis. I am paid to understand my craft not
how to manipulate a program such as Excel. I don't mean to vent on the MVP's
as they are relaying the information regarding the program and I appreciate
their efforts. Excel has now been reduced to a simple number cruncher not
worth any type of detailed analysis required in engineering in my view.
Click to expand...

It never was, no matter how many cool hacks people managed to wedge into
Excel.
With free stuff like Octave, SciLab, and R, I can't understand why
people try to do engineering in a spreadsheet program.

My famous counterexample as to why NOT to use Excel:

http://home.comcast.net/~cgwcgw/pong.zip
 
J

JE McGimpsey

Gregg Larson said:
As I only use the data analysis functions on occasion I find that the removal
of this function totally without basis.
Click to expand...

You misunderstand. The removal of the wizards has nothing to do with how
frequently you use it, nor with any technical decision about the merits
of the wizards themselves. The "basis" for removing them is that their
implementation was dependent on *other* features which were removed, not
on a whim, but because of the inability to implement them in the
required time with the available resources.
I am paid to understand my craft not how to manipulate a program such
as Excel. I don't mean to vent on the MVP's as they are relaying the
information regarding the program and I appreciate their efforts.
Excel has now been reduced to a simple number cruncher not worth any
type of detailed analysis required in engineering in my view.
Click to expand...

Hmmm...

As an engineer, your premise seems curious, to say the least.

If I use a tool only infrequently, it's incumbent upon me to ensure that
I know not only how to use it, but also what its limits and boundary
conditions are (as well as that it's giving a correct answer within
those conditions). That is, how to manipulate it.

This is especially true for functions and wizards stemming from an
add-in with a history rife with errors and poor numerical approximations.

If one were to use Excel in engineering applications (or any other
application where one is being paid, for that matter) *without* having
done a detailed analysis, that borders on malpractice, in my view.

When performing a peer review, I would *never* sign off on a solution
generated by a wizard unless I was able to trace the XL functions that
the wizard used (having previously validated those functions).

Fortunately, most of the frequently used wizards simply use XL functions
to create a pretty result - something that can be easily replicated by
the Oatbran and similar workbooks that have been referenced in these
threads.

The situation with Solver is a bit different, as it manipulates
functions in target cells, which can then be evaluated on their own. The
loss of that ability hurts badly. Still, I've seen more than my share of
Solver solutions where a lack of understanding of the nature of the
problem produced non-optimal solutions that were justified with "but
that's what Solver came up with".
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Similar Threads

Cell data entering slow! 0
ADD-INS 3
Data Analysis Tab 1
Regression Analysis on Excel 2
Missing Tools/Data Analysis on the horribleMicrosoft Office 2008. Any help? 1
Data Analysis Tool 1
Solver and data analysis Add in 1
Correlation tool says Input range must be contiguous reference. 1

Top