1 / 2

2.5 The coefficient of determination (r 2 ) & correlation (r)

2.5 The coefficient of determination (r 2 ) & correlation (r) ¿How good is the estimated regression line? Since residuals are (+) & (-), that is, the estimated values are different than those observed  we should get an indication of the “goodness of fit”

moral
Download Presentation

2.5 The coefficient of determination (r 2 ) & correlation (r)

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. 2.5 The coefficient of determination (r2) & correlation (r) ¿How good is the estimated regression line? Since residuals are (+) & (-), that is, the estimated values are different than those observed  we should get an indication of the “goodness of fit” The r2 measures the share (or %) of the total change in “Y” which is explained by the regression model (the “Xs”) We have: 0 =< r2 =< 1 ; no fit v. perfect fit Moreover: r is the sq. root of r2, but in regression analysis has no direct interpretation  -1 =< r=< 1

  2. 2.5 The coefficient of determination (r2) & correlation (r) Other ways to check the goodness of fit: • Variance / std. error of the model • Joint hypotheses test (F-test) These will serve for model selection, assessing the variability in the observed “Y” in relation to those explained In addition, the variances (& std. errors) of the estimated coefficients give us an idea of the variability of the coefficients from sample to sample Given GM  with CLRM & LS the variability is minimum

More Related