Determining the distribution of fitness effects using a generalized Beta-Burr distribution

Paul Joyce and Zaid Abdo

Theoretical Population Biology, 2018, vol. 122, issue C, 88-96

Abstract: In Beisel et al. (2007), a likelihood framework, based on extreme value theory (EVT), was developed for determining the distribution of fitness effects for adaptive mutations. In this paper we extend this framework beyond the extreme distributions and develop a likelihood framework for testing whether or not extreme value theory applies. By making two simple adjustments to the Generalized Pareto Distribution (GPD) we introduce a new simple five parameter probability density function that incorporates nearly every common (continuous) probability model ever used. This means that all of the common models are nested. This has important implications in model selection beyond determining the distribution of fitness effects. However, we demonstrate the use of this distribution utilizing likelihood ratio testing to evaluate alternative distributions to the Gumbel and Weibull domains of attraction of fitness effects. We use a bootstrap strategy, utilizing importance sampling, to determine where in the parameter space will the test be most powerful in detecting deviations from these domains and at what sample size, with focus on small sample sizes (n<20). Our results indicate that the likelihood ratio test is most powerful in detecting deviation from the Gumbel domain when the shape parameters of the model are small while the test is more powerful in detecting deviations from the Weibull domain when these parameters are large. As expected, an increase in sample size improves the power of the test. This improvement is observed to occur quickly with sample size n≥10 in tests related to the Gumbel domain and n≥15 in the case of the Weibull domain.

Keywords: Generalized Beta-Burr distribution; Distribution of fitness effects; Extreme value theory; Adaptive mutation distribution; Likelihood ratio testing (search for similar items in EconPapers)
Date: 2018
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0040580917300357
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:thpobi:v:122:y:2018:i:c:p:88-96

DOI: 10.1016/j.tpb.2017.07.001

Access Statistics for this article

Theoretical Population Biology is currently edited by Jeremy Van Cleve

More articles in Theoretical Population Biology from Elsevier
Bibliographic data for series maintained by Catherine Liu ().