In bivariate analysis, is there an option for handling outliers that may lead to incorrect estimates?
Feb 24, 2016 6:13 AM
Yes. Jerry Cooper demonstrates that if you identify outliers after fitting the data (Analyze>Fit Y by X), you can select the Fit Robust option from the resulting Bivariate Fit red triangle (Robust>Fit Robust). You can compare the Robust Fit to any other Fit options by viewing your selections on the resulting graph and statistics. The Robust option provides two methods to reduce the influence of outliers in your data set: Robust Fit and Cauchy Fit. See Robust.