I have a regression equation in which I am curious if the amount of scatter around the regression line is driven by phylogeny. I know there is a lot of phylogenetic signal because Pascal's lambda is very high, but I am more interested in seeing how much it affects the strength of the relationship between X and Y. I had thought about trying to calculate phylogenetic independent contrasts and using those to see if the spread about the regression line was less under phylogenetic independent contrasts than data that does not take phylogeny into account, but in thinking about it I realized that I haven't seen phylogenetic contrasts done in biology in what seems like ages. Everyone seems to be doing Phylogenetic Generalized Least Squares now, and if I recall right I remember someone like Revell (2010) mentioning that PGLS and independent contrasts are functionally identical.
My question is, do people still use phylogenetic independent contrasts in biology and if so would this be an appropriate use of them to investigate my data.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression