Re: Why is it "least squares" and not "least absolute diffrerence"?
- From: "Dann Corbit" <dcorbit@xxxxxxxxx>
- Date: Thu, 31 Jul 2008 00:26:51 -0700
"Lurfys Maw" <invalid@xxxxxxxxxxxxxxx> wrote in message news:jjk2949aog5ag5517njt6vuujqedr3jc4b@xxxxxxxxxx
Why is the "least squares" method used in calculating a regression
value rather than a "least absolute differences"?
Why are the differences squared? Why not take the absolute values of
the differences and sum those?
Look at the first paragraph of this:
http://en.wikipedia.org/wiki/Least_squares
** Posted from http://www.teranews.com **
.
- References:
- Why is it "least squares" and not "least absolute diffrerence"?
- From: Lurfys Maw
- Why is it "least squares" and not "least absolute diffrerence"?
- Prev by Date: free to join earn money easy
- Next by Date: Re: Histogram
- Previous by thread: Why is it "least squares" and not "least absolute diffrerence"?
- Next by thread: Re: Why is it "least squares" and not "least absolute diffrerence"?
- Index(es):
