Estimating Ancestral Divergence Times Using Lognormal Distribution
Author Information
Author(s): David A. Morrison
Primary Institution: National Veterinary Institute and Swedish University of Agricultural Sciences, Uppsala, Sweden
Hypothesis
Divergence times should be presented using geometric means rather than arithmetic means to improve accuracy.
Conclusion
Using geometric means for estimating divergence times can lead to more accurate biological conclusions compared to arithmetic means.
Supporting Evidence
- Simulations showed that arithmetic averages can overestimate divergence times by up to 35%.
- Geometric means provide a better fit for the distribution of divergence times in most cases.
- Using geometric means can lead to different biological conclusions than using arithmetic means.
Takeaway
This study shows that when scientists estimate how long ago species split from each other, they should use a special kind of average called the geometric mean instead of the regular average to get better results.
Methodology
The study involved computer simulations and empirical data analyses to compare the effectiveness of geometric means versus arithmetic means in estimating divergence times.
Potential Biases
Potential biases may arise from insufficient taxon sampling and the quality of fossil calibrations.
Limitations
The study primarily focuses on the precision of estimates rather than their accuracy, and it does not address all potential sources of bias.
Want to read the original?
Access the complete publication on the publisher's website