hi Ecolog, I was wondering if anyone had any good how-to references on comparing fitted lines. Here is what I usually do, particularly if there is one x and two y's paired.
1. Save predicted values from fit line and do a t-test on the two arrays. 2. Save residuals and examine patterns between fit models. 3. Include confidence intervals for fit lines, and they don't overlap, the lines are different. 4. Use resampling to examine the likelihood that these two fit lines would emerge from a population of points. 5. Take difference between every pair of response points (provided share independent variable), and use a t-test to determine if the mean is different from 0. Is there a best/preferred or standard approach that use in ecology? cheers, chris lortie.
