Fundamentals of Statistics contains material of various lectures and courses of H. Lohninger on statistics, data analysis and chemometrics......click here for more. 
See also: variable selection, Data Set  Gasoline Samples  
Mallows Cp Statistic
C.L. Mallows developed a method to find adequate models by plotting a special statistic against the number of variables+1. C_{p} = SS_{res}/MS_{res}  N + 2p, SS_{res} is the residual sum of squares for the model with p1 variables, MS_{res} is the residual mean square when using all available variables, N is the number of observations, and p is the number of variables used for the model plus one. The general procedure to find an adequate model by means of the C_{p} statistic is to calculate C_{p} for all possible combinations of variables and the C_{p} values against p. The model with the lowest C_{p }value approximately equal to p is the most "adequate" model.


Last Update: 20121008