Re: Usage of the regression subroutines

From: Dennis Shea <shea_at_nyahnyahspammersnyahnyah>
Date: Wed, 11 Jan 2006 17:44:35 -0700 (MST)

>I have some question about the usage of the regression subroutines "regline"
>and "regcoef". Once we got the regression coefficient between X and Y, how
>we get Y_hat(the predicted value of y) based on the linear regression model
>Y_hat=Beta0 + Beta1*X. I can identify Beta1 is the return value of the
>subroutines above. Where can I find anything about Beta0?


  Sometime ago the "yintercept" attribute was added to regline.
  The documentation was not updated to reflect this. Sorry.
        yReg = rc*x + rc_at_yintercept
  Please see Example 1
  Note: the y-intercept [x=0] could be calculated from the other
        returned attributes.
        yReg = rc*(x-rc_at_xave) + rc_at_yave
        yintercept = 0.974561*(0-2165.0) + 2125.278 = 15.35229
                       rc x=0 xave yave

   Well, as can be ascertained by the function name "regCoef",
   it was meant to return the regression coefficient only.
   The 3 quantities we thought would be useful for statistical
   evaluation: tval, rstd and npts were also returned.
   I guess we could include the yintercept in a future release.
   Comment: the yintercept is just a constant that shifts the
            estimated value up or down on an x-y plot.
            My experience with usage of the regression coef is as follows.
            Note in the above that (x-rc_at_xave) is an anomaly (say, xAnom)
            You could get the projection of the regression coeffients
            upon the data by using the anomalies of the "x" grid
                yProjection = rc*xAnom

ncl-talk mailing list
Received on Wed Jan 11 2006 - 17:44:35 MST

This archive was generated by hypermail 2.2.0 : Thu Jan 12 2006 - 09:08:16 MST