[EMAIL PROTECTED] wrote: > On 11 Feb 2004 at 11:40, Rajarshi Guha wrote: > > Leverage is kind of Mahalanobis-distance in X-space, so make > sense also for observations not used to estimate the model. (But > maybe not the name!). High leverage for this points indicates that it > is far from the points used to estimate the model, so you have really > to trust the form of the model! > > Kjetil Halvorsen
Yes but please don't confuse high leverage with high influence. A point can have high leverage and low influence, and vice versa. Observations not used in the estimation of the model may indeed be far away from the center of the X-space (high-leverage) but they have zero influence on the model. A better way to determine how well predicted an observation will be (assuming the data point was not used in the estimation of the model), as suggested by Don Burrill, is to compute the variance of the prediction. -- Paige Miller Eastman Kodak Company paige dot miller at kodak dot com http://www.kodak.com "It's nothing until I call it!" -- Bill Klem, NL Umpire "When you get the choice to sit it out or dance, I hope you dance" -- Lee Ann Womack >> I know that when I create a multiple linear regression model I can >> evaluate leverage values for the points used to create the model. >> >>From what I understand leverage gives an indication of how much a >>given observation affects the coefficients in the model. Thus when I >>use the model to predict values for some unknown observations (ie >>observations not used to create the model) calculating leverage values >>for these observations does not make sense - is this correct? >> >>If the above is true is there any measure that I could use (apart from >>R2 values or residuals) that I can use to obtain some indication of >>how the good the coefficients are when predicting unknown observations >>(sort of a 'reverse leverage' value)? . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
