If the graph is linear then it is a straight line and y=a+bx is the equation of a straight line...so I can't see the problem ....Am I mising something???
Paul
> -----Original Message-----
> From: Jeremy Miles [SMTP:[log in to unmask]]
> Sent: 15 June 2001 11:06
> To: [log in to unmask]
> Subject: Re: Correlation
>
> At 21:10 14/06/01, you wrote:
> >A friend of mine has asked for help with the following problem:
> >
> >Suppose a scatter diagram has perfect positive correlation and the graph is
> >linear. Is there a proof that the equation of the line is y = a + bx?
>
> Hi Paul (and list),
>
> Its a tricky one to see exactly what you mean. I would guess that one
> approach would be to calculate the predicted values, based on the equation
> y = a + bx, and then plot the values of y against the predicted values,
> based on the equation.
>
> The plot would be a straight line, and the correlation (if you calculated
> it) would be 1.00, showing that the equation gives perfect prediction of
> the values of y in the data.
>
> However, you are then (re) using the equation, and are stuck in a circular
> argument, because proving that y = a + bx requires believing that y = a + bx.
>
> Maybe that will help anyway,
>
> JM
>
>
>
>
>
>
>
>
> ----------
> Dr Jeremy Miles
> [log in to unmask]
> Phone: 01332 592090, Fax: 01332 593131, Mobile: 07941 228018
> Inst. of Behavioural Sciences, Derby University, Derby, DE22 3HL, UK
> http://ibs.derby.ac.uk/~jeremym / http://www.jeremymiles.co.uk
>
> NOTE: After 1st August, 2001, Department of Health Sciences, University of
> York, Heslington, York, YO10 5DD
|