Stepwise feature selection

From Knowledge Discovery

from the matlab statistics toolbox

function outF=stepwise(X,y,in,penter,premove)

STEPWISE Interactive tool for stepwise regression.

  STEPWISE(X,Y) displays an interactive tool for creating a
  regression model to predict the vector Y using a subset of the
  predictors given by columns of the matrix X.  Initially no
  predictors are included in the model, but you can click on
  predictors to switch them into and out of the model.  STEPWISE
  automatically includes a constant term in all models.
  .
  For each predictor in the model, its least squares coefficient
  is plotted with a blue filled circle.  For each predictor not
  in the model, a filled red circle indicates the coefficient it
  would have if it were added to the model.  Horizontal bars indicate
  90% (colored) and 95% (black) confidence intervals.

  STEPWISE(X,Y,INMODEL,PENTER,PREMOVE) specifies the initial state
  of the model and the confidence levels to use.  INMODEL is a logical
  or index vector specifying the predictors that should be in the
  initial model (default is none).  PENTER specifies the maximum
  p-value for a predictor to be recommended for adding to the model
  (default 0.05).  PREMOVE specifies the minimum p-value for a

Copyright 1993-2005 The MathWorks, Inc.