Published September 28, 2023 | Version v1
Journal article Open

Analytical advances alleviate model misspecification in non--Brownian multivariate comparative methods

Creators

  • 1. Department of Computer and Information Science, Linköping University, Linköping, Sweden
  • 2. Department of Biological Sciences and the Institutes of Environment, Florida International University, Miami, Fl, 33199, USA
  • 3. IntiQuan GmbH, Basel, Switzerland
  • 4. Department of Ecology and Biogeography, Nicolaus Copernicus University in Toruń, Toruń, Poland
  • 5. nstitute of Evolutionary Biology, Faculty of Biology, Biological and Chemical Research Centre, University of Warsaw
  • 6. Natural History Museum, University of Oslo, Oslo, Norway

Description

Adams and Collyer (2018) argue that contemporary multivariate (Gaussian) phylogenetic comparative methods are prone to favouring more complex models of evolution and sometimes rotation invariance can be an issue. Here we dissect the concept of rotation invariance and point out that, depending on the understanding, this can be an issue with any method that relies on numerical instead of analytical estimation approaches. We relate this to the ongoing discussion concerning phylogenetic principal components analysis.
Contrary to what Adams and Collyer (2018) found, we do not observe a bias against the simpler Brownian motion process in simulations when we use the new, improved, likelihood evaluation algorithm employed by mvSLOUCH, which allows for studying much larger phylogenies and more complex model setups.

Notes

Additional (potentially updated) simulation results and data can be found at https://github.com/krzbar/KJVJMRKK_mvSLOUCH . Furthermore, connected data and simulation results can be found at https://doi.org/10.5061/dryad.sj3tx9656 and https://doi.org/10.5281/zenodo.7499740 .

Files

BartoszekGonzalezMitovPienaarPiwczynskiPuchalkaSpalikVoje_RotationInvariancePCMs_code.zip

Files (862.5 MB)