[TI-M] Re: Statistics
[Prev][Next][Index][Thread]
[TI-M] Re: Statistics
_______________________________________________________________
Sign up for FREE iVillage newsletters <http://s.ivillage.com/rd/16705> .
>From health and pregnancy to shopping and relationships, iVillage
has the scoop on what matters most to you.
<-----Original Message----->
>
>From: jeff
>Sent: 5/22/2002 5:05:49 PM
>To: ti-math@lists.ticalc.org
>Subject: [TI-M] Re: Statistics
>
>
>I learned a formula for linear regression in an engineering class
freshman year of high school,
>but seem to have forgotten it. It involved a long series of sums of
x's, y's, and various squared
>terms i think... Hmm... Here, I looked this up and think it's right:
>For y=m*x+b: (using E instead of sigma for sum)
> Ey * E(x^2) - Ex * E(xy)
>m = ------------------------
> n * E(x^2) - (Ex)^2
>
> n * E(xy) - Ex * Ey
>b= -------------------
> n * E(x^2) - (Ex)^2
>
>Those look familiar; I think they're right. Not going to check them. I
got them from
>mathworld.com. They don't seem to have any other regressions there. If
anyone has ever hacked
>into the calc OS, they'd probably be able to find the algorithms...but
I don't really know much
>about that...
>
>Jeffrey
__________________________________________________
Actually, if you go to STATS|MATH you will find linear, quadratic,
cubic, quartic, logarithmic, power, and exponential regression.
KrazyKyngeKorny
Follow-Ups: