MathGroup Archive 2005

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: LinearRegression

  • To: mathgroup at smc.vnet.net
  • Subject: [mg60023] Re: LinearRegression
  • From: "Frank Iannarilli" <frankeye at cox.net>
  • Date: Sun, 28 Aug 2005 23:14:34 -0400 (EDT)
  • References: <dehjfi$c4p$1@smc.vnet.net>
  • Sender: owner-wri-mathgroup at wolfram.com

Needs["Statistics`MultinormalDistribution`"]

mu = {0, 0};

Here's your correlation coefficient
r = 0.99;
covar = {{1, r}, {r, 1}};

numpts=10;
pts = RandomArray[MultinormalDistribution[mu, covar], numpts];

Here's your y=mx+b parameters m and b:
m = 4.; b = 10;

scaledpts = Map[({#[[1]], #[[2]]*m + b}) &, pts];

ListPlot[scaledpts];

The list "pts" of {x,y} points will be (on the average) centered about
{0,0}, with equal spread of x values and y values. Moreover, they will
exhibit the desired correlation coefficient r.

The list "scaledpts" simply scales and shifts the y-values per the
desired mx+b fit.  

Regards,


  • Prev by Date: Combinatorica: Edge Options
  • Next by Date: Re: Keyboard problems - no "dead keys" - in Linux version of Mathematica
  • Previous by thread: LinearRegression
  • Next by thread: Re: LinearRegression