?

Log in

No account? Create an account
 
 
15 February 2009 @ 04:40 pm
I think this is a statistics question, but I'm not sure  
I am somewhat known, around the university, for being able to make Microsoft Excel do things. And I've been asked to do a consult next week.

One of the questions is, can you do "least squares" in Excel? Since I'm not entirely certain what "least squares" is, I don't know. (I looked it up in the help file and didn't find it, but there's probably an add-on pack that'll do it, if I know what to look for.)

Help?
Tags:
 
 
 
Villiersdianavilliers on February 15th, 2009 09:50 pm (UTC)
Yes you can

Try searching for linear regression or failing that slope and intercept. (I presume they're trying to fit a straight line.)
Janet Miles, CAP-OMjanetmiles on February 15th, 2009 10:19 pm (UTC)
Thanks!
guppiecatguppiecat on February 15th, 2009 09:51 pm (UTC)
least square is a method of fitting a curve to a scatter plot of data. You can do it, but you need the data analysis add in (at least, you used to).

See http://pages.pomona.edu/~wes04747/lab/least_sq.doc
Janet Miles, CAP-OMjanetmiles on February 15th, 2009 10:21 pm (UTC)
That looks helpful. Thanks!
Angela di Tenebreditenebre on February 15th, 2009 09:53 pm (UTC)
There is a "least squares" page in Wikipedia that will give you some idea of what you're dealing with, though, this may be what you're looking for:

"Constructing a Least-Squares Graph Using Microsoft Excel"
http://www.chem.purdue.edu/gchelp/tools/graphing.html
Janet Miles, CAP-OMjanetmiles on February 15th, 2009 10:22 pm (UTC)
That's probably what I'm looking for. Thanks!
n5red on February 15th, 2009 10:20 pm (UTC)
You're an expert at Microsoft Excel? Somehow I had a much higher opinion of you. It's just so sad to see such an outstanding person go so wrong. It really breaks my heart.
Janet Miles, CAP-OMjanetmiles on February 15th, 2009 10:23 pm (UTC)
Honestly, I think Excel is the single best thing ever to come out of Redmond.

If it helps at all, I used to be very good with Lotus 1-2-3 and Quattro Pro. :-)
Cymru Llewes: profilecymrullewes on February 16th, 2009 02:02 am (UTC)
I can agree with that. However someone I worked with had this opinion about M$ Access being much more powerful than Excel.
Janet Miles, CAP-OM: academiajanetmiles on February 16th, 2009 02:21 am (UTC)
Access is more powerful than Excel in a lot of ways, but I don't like it as much.
gh4acws on February 16th, 2009 03:49 pm (UTC)
apples and oranges
spreadsheet vs. database frontend.

Admittedly one CAN sore data in a spreadsheet - but usually without the sensible restrictions tat a database requires [ a halfway sensibly built database will ensure that John Smith is not employee and not-employee at the same time { to avoid sexist male AND female }, that you own the first season of BSG or not, not both own and not own. ] to take a set of numbers though and do fast operations on them - turnover by calendar-week or branches and sums of one branch over all weeks or all branches in one week : that is a strength of Excel - specially if the Poughkeepsie branch always delivers its numbers late and has to correct them twice. One just slots in their updated numbers in the right fields and gets all the updated results.

Both excel and access are terrible if one simply wants to write text - and for doing accounting Word is not the software I would choose.
All this apart from certain suspicion of anything made in Redmont.
gh4acws on February 16th, 2009 03:52 pm (UTC)
sorry for the spelling
if one is tired even simple websoftware is too complex.
Cymru Llewes: profilecymrullewes on February 16th, 2009 08:57 pm (UTC)
Re: apples and oranges
I actually like "writing" in Excel but that is because I'm generally taking notes. And all of my notes need to correlate to certain person or object or date.