Table 1 

Loglikelihoods of a linear fit between all ω values and each of 8 common distributions, with likelihood ratio tests for the differences in distributions calculated for the 3 best distributional fits 

Distribution 
Pearson's r^{a} 
k^{b} 
LRT^{c} 
df 
Pvalue^{d} 


Weibull 
.999 
2 
186.39 
2 
<10^{6} 
Gamma 
.999 
2 
201.67 
2 
<10^{6} 
Exponential 
.998 
1 
28.29 
1 
<10^{6} 
Logistic 
.924 
2 
^{e} 
 
 
Normal 
.923 
2 
 
 
 
Extreme Value 
.903 
3 
 
 
 
LogNormal 
.854 
2 
 
 
 
Cauchy 
.163 
2 
 
 
 


a. Pearson's r is the linear correlation of the data to the quartiles, based on the maximum likelihood inferred parameters for each family of distributions. b. k is the number of free parameters in the distribution. c. Likelihood ratio test: LRT = 2 * (loglikelihood of metabolic ω values + loglikelihood of nonmetabolic ω values)  (loglikelihood for all ω values). d. Distributed χ^{2}. e. Likelihood ratio test only performed for the best distributional fits. 

Hudson and Conant BMC Evolutionary Biology 2011 11:89 doi:10.1186/147121481189 