I am using non-linear regression to fit electophysiological data (current vs t) to exponential equations. I am using an F-test on the residual sum of squares to determine how many components are required. A typical trace will have several thousand points. Question: If I use an adjacent average technique to smooth the data, (i.e keeping the number of points the same) am I justified in using the same degrees of freedom for the F test as I would with the raw data (Number of points -number of parameters)? Thnak you. ================================================================= Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =================================================================