Bayesian Quantile Regression for Longitudinal Studies with Nonignorable Missing Data

Summary .  We study quantile regression (QR) for longitudinal measurements with nonignorable intermittent missing data and dropout. Compared to conventional mean regression, quantile regression can characterize the entire conditional distribution of the outcome variable, and is more robust to outliers and misspecification of the error distribution. We account for the within-subject correlation by introducing a   ℓ₂   penalty in the usual QR check function to shrink the subject-specific intercepts and slopes toward the common population values. The informative missing data are assumed to be related to the longitudinal outcome process through the shared latent random effects. We assess the performance of the proposed method using simulation studies, and illustrate it with data from a pediatric AIDS clinical trial.     

ABC as a flexible framework to estimate demography over space and time: some cons,many pros

The analysis of genetic variation to estimate demographic and historical parameters and to quantitatively compare alternative scenarios recently gained a powerful and flexible approach: the Approximate Bayesian Computation (ABC). The likelihood functions does not need to be theoretically specified, but posterior distributions can be approximated by simulation even assuming very complex population models including both natural and human‐induced processes. Prior information can be easily incorporated and the quality of the results can be analysed with rather limited additional effort. ABC is not a statistical analysis per se, but rather a statistical framework and any specific application is a sort of hybrid between a simulation and a data‐analysis study. Complete software packages performing the necessary steps under a set of models and for specific genetic markers are already available, but the flexibility of the method is better exploited combining different programs. Many questions relevant in ecology can be addressed using ABC, but adequate amount of time should be dedicated to decide among alternative options and to evaluate the results. In this paper we will describe and critically comment on the different steps of an ABC analysis, analyse some of the published applications of ABC and provide user guidelines.     

Computerized matching of relevés and association tables,with an application to the British National Vegetation Classification

When a new relevé is to be assigned to a pre-existing type, its composition is compared with an association table. Bayesian inference may seem a good way to make the comparison, but presents difficulties. In an alternative approach, three indices of goodness-of-fit are proposed. Compositional satisfaction is a measure of how well the species composition of the relevé fits the constancy classes in the table; it is a minor modification of the Czekanowski coefficient of similarity between observed and expected numbers of species in each constancy class. Dominance satisfaction is a modification of the Czekanowski similarity between the relevé and cover values that might be expected from the association table. Dominance constancy is a weighted mean of the constancy class of the four most abundant species in the relevé. A computer program, TABLEFIT, combines them into a single index. It has been tested on British mire vegetation.     

The posterior probability distribution of alignments and its application to parameter estimation of evolutionary trees and to optimization of multiple alignments

How to sample alignments from their posterior probability distribution given two strings is shown. This is extended to sampling alignments of more than two strings. The result is first applied to the estimation of the edges of a given evolutionary tree over several strings. Second, when used in conjunction with simulated annealing, it gives a stochastic search method for an optimal multiple alignment.Correspondence to: L. Allison     

Comparisons of dipteran, hymenopteran and coleopteran parasitoids: provisional phylogenetic explanations

We consider differences between dipteran, hymenopteran and coleopteran parasitoids in the following categories: taxonomic range and developmental stage of hosts attacked; habitats they are attacked in; developmental stage of the parasitoid contacting the host; occurrence of phoresy, and attacking hosts during flight. Using existing phylogenetic classifications we reconstruct possible ancestral conditions to the parasitoid clades in the three orders. By considering these as phylogenetic constraints and potentialities we attempt to account for the observed differences between the parasitoids within the orders.     

Causal inference and large-scale expert validation shed light on the drivers of SDM accuracy and variance

Aim

To develop a causal understanding of the drivers of Species distribution model (SDM) performance.

Location

United Kingdom (UK).

Methods

We measured the accuracy and variance of SDMs fitted for 518 species of invertebrate and plant in the UK. Our measure of variance reflects variation among replicate model fits, and taxon experts assessed model accuracy. Using directed acyclic graphs, we developed a causal model depicting plausible effects of explanatory variables (e.g. species' prevalence, sample size) on SDM accuracy and variance and quantified those effects using a multilevel piecewise path model.

Results

According to our model, sample size and niche completeness (proportion of a species' niche covered by sampling) directly affect SDM accuracy and variance. Prevalence and range completeness have indirect effects mediated by sample size. Challenging conventional wisdom, we found that the effect of prevalence on SDM accuracy is positive. This reflects the facts that sample size has a positive effect on accuracy and larger sample sizes are possible for widespread species. It is possible, however, that the omission of an unobserved confounder biased this effect. Previous studies, which reported negative correlations between prevalence and SDM accuracy, conditioned on sample size.

Main conclusions

Our model explicates the causal basis of previously reported correlations between SDM performance and species/data characteristics. It also suggests that niche completeness has similarly large effects on SDM accuracy and variance as sample size. Analysts should consider niche completeness, or proxies thereof, in addition to sample size when deciding whether modelling is worthwhile.