R: I am starting to do LOESS regressions and was wondering if anyone has written code to create M-Plots (Cleveland and Devlin 1988). The M-Plot allows a method for choosing the smoothing parameter that produces the smallest variance subject to not generating a statistically significant bias in the fitted values; thus you can do anova on loess fits. Cleveland, William S., Susan J. Devlin (1988), "Locally Weighted Regression: An Approach to Regression Analysis by Local Fitting," Journal of the American Statistical Association 83, 596-610. Thanks, Jason Higbee Research Associate Federal Reserve Bank of St. Louis [[alternative HTML version deleted]]