Data from: Detecting adaptive evolution in phylogenetic comparative analysis using the Ornstein-Uhlenbeck model

Phylogenetic comparative analysis is an approach to inferring evolutionary process from a combination of phylogenetic and phenotypic data. The last few years have seen increasingly sophisticated models employed in the evaluation of more and more detailed evolutionary hypotheses, including adaptive hypotheses with multiple selective optima and hypotheses with rate variation within and across lineages. The statistical performance of these sophisticated models has received relatively little systematic attention, however. We conducted an extensive simulation study to quantify the statistical properties of a class of models toward the simpler end of the spectrum that model phenotypic evolution using Ornstein-Uhlenbeck processes. We identify key determinants of statistical power in model selection. We find also that model parameter estimates are inherently difficult to estimate accurately, indicating a relative paucity of information in the data relative to these parameters. We therefore recommend that investigators explore the precision of their estimates through resampling methods such as the parametric bootstrap, before basing conclusions on parameter estimates. We argue that weak identifiability of parameter estimates need not forestall meaningful inference based on model selection. Inasmuch as more sophisticated methods include these models as special cases, our results have implications for these more parameter-rich methods. To unsubscribe from this group and stop receiving emails from it, send an email to [e-mail].系统发育比较分析（phylogenetic comparative analysis）是一种结合系统发育数据与表型数据以推断进化过程的研究方法。近年间，用于评估愈发精细的进化假说的模型日趋复杂精密——此类假说涵盖了带有多重选择最优值的适应性假说，以及谱系内部与跨谱系的进化速率变异假说。然而，这类复杂模型的统计性能却鲜少得到系统性的关注。本研究开展了大规模模拟实验，以量化一类处于模型复杂度谱系偏简易一端的模型的统计特性，这类模型采用奥恩斯坦-乌伦贝克（Ornstein-Uhlenbeck）过程模拟表型进化。本研究明确了模型选择过程中统计功效的关键决定因素；同时发现，模型参数估计本身难以实现精准拟合，这表明相较于待估参数而言，数据集所蕴含的信息相对匮乏。因此，研究者在基于参数估计得出结论前，应通过参数自助法（parametric bootstrap）等重采样方法对估计结果的精度进行探究。本研究认为，参数估计的弱可识别性并不妨碍基于模型选择开展有意义的统计推断。鉴于更为复杂的模型方法均将此类模型作为其特例纳入考量，本研究结果对这类高参数模型方法同样具有参考价值。若要退订该群组并停止接收相关邮件，请发送邮件至[e-mail]。