A general approach for two-stage analysis of multilevel clustered non-Gaussian data

In this article, we propose a two-stage approach to modeling multilevel clustered non-Gaussian data with sufficiently large numbers of continuous measures per cluster. Such data are common in biological and medical studies utilizing monitoring or image-processing equipment. We consider a general class of hierarchical models that generalizes the model in the global two-stage (GTS) method for nonlinear mixed effects models by using any square-root-n-consistent and asymptotically normal estimators from stage 1 as pseudodata in the stage 2 model, and by extending the stage 2 model to accommodate random effects from multiple levels of clustering. The second-stage model is a standard linear mixed effects model with normal random effects, but the cluster-specific distributions, conditional on random effects, can be non-Gaussian. This methodology provides a flexible framework for modeling not only a location parameter but also other characteristics of conditional distributions that may be of specific interest. For estimation of the population parameters, we propose a conditional restricted maximum likelihood (CREML) approach and establish the asymptotic properties of the CREML estimators. The proposed general approach is illustrated using quartiles as cluster-specific parameters estimated in the first stage, and applied to the data example from a collagen fibril development study. We demonstrate using simulations that in samples with small numbers of independent clusters, the CREML estimators may perform better than conditional maximum likelihood estimators, which are a direct extension of the estimators from the GTS method.     

An ecological example of the application of projection pursuit to compositional data

A modern statistical technique for the exploratory analysis of multivariate data, called projection pursuit, is described. The method consists of the extraction and display of informative one- or two-dimensional projections of the multivariate dataset. Projection pursuit has much scope for application to ecological datasets. The example in this paper concerns element concentrations in soil samples from a coastal embayment on the east coast of Australia. The application attempts to search for patterns in the soils which could be related to vegetation patterns. The compositional nature of the dataset adds a novel element to the methodology.     

Network-based diffusion analysis: a new method for detecting social learning

Social learning has been documented in a wide diversity of animals. In free-living animals, however, it has been difficult to discern whether animals learn socially by observing other group members or asocially by acquiring a new behaviour independently. We addressed this challenge by developing network-based diffusion analysis (NBDA), which analyses the spread of traits through animal groups and takes into account that social network structure directs social learning opportunities. NBDA fits agent-based models of social and asocial learning to the observed data using maximum-likelihood estimation. The underlying learning mechanism can then be identified using model selection based on the Akaike information criterion. We tested our method with artificially created learning data that are based on a real-world co-feeding network of macaques. NBDA is better able to discriminate between social and asocial learning in comparison with diffusion curve analysis, the main method that was previously applied in this context. NBDA thus offers a new, more reliable statistical test of learning mechanisms. In addition, it can be used to address a wide range of questions related to social learning, such as identifying behavioural strategies used by animals when deciding whom to copy.     

A new genus of Aeolothripidae (Thysanoptera) from New Zealand and New Caledonia

Desmidothrips n.gen. is described from New Zealand, with walkerae n.sp. as type‐species. Aeolothrips inauditus Bianchi from New Caledonia is transferred to this new genus. Desmidothrips is related to the Australian genus Desmothrips, and is the only member of the Aeolothripidae native to New Zealand.     

Estimation of Black Bear Abundance Using a Discrete DNA Sampling Device

Abstract We developed a snare for collection of black bear (Ursus americanus) hair that obtained a unique hair sample at each snare site, improved the quantity of collected hair compared to barbed-wire corrals, and was easy to deploy over a wide range of topographical features and habitat conditions. This device allowed us to implement intensive sampling methodology needed in mark-recapture experiments with minimal effort. By improving the quantity of hair collected, we also lowered the potential for bear identification errors at the lab. During 2003–2004, bears in 2 study areas triggered snares 1,104 times, which resulted in the collection of 981 hair samples. Of the samples we collected, 79% (775) produced valid genetic data. In 2003, 454 samples identified 79 genetically distinct individuals, and 321 samples identified 86 genetically distinct individuals in 2004. Analysis of capture-recapture data indicated that capture probabilities were affected by heterogeneity among individuals and behavioral responses, but showed little evidence of time effects. Consequently, we used the Pollock and Otto (1983) estimator for model M_bh to estimate abundance with reasonably good precision (CV: 12–14%). Density on the Steamboat and Toketee, Oregon, USA, study areas over the 2-year period averaged 19 bears/100 km² and 22 bears/100 km², respectively. Average capture and recapture probabilities over the 2 years of the study were 30% and 63%, respectively, indicating a trap-prone behavioral response. Knowledge of bear densities on the Steamboat and Toketee study areas will enable managers to set hunting quotas, advise land management agencies on habitat issues, and create a baseline database to assist in the long-term monitoring of bear trends in a changing landscape.     

Blinded and unblinded sample size reestimation procedures for stepped‐wedge cluster randomized trials

The ability to accurately estimate the sample size required by a stepped‐wedge (SW) cluster randomized trial (CRT) routinely depends upon the specification of several nuisance parameters. If these parameters are misspecified, the trial could be overpowered, leading to increased cost, or underpowered, enhancing the likelihood of a false negative. We address this issue here for cross‐sectional SW‐CRTs, analyzed with a particular linear‐mixed model, by proposing methods for blinded and unblinded sample size reestimation (SSRE). First, blinded estimators for the variance parameters of a SW‐CRT analyzed using the Hussey and Hughes model are derived. Following this, procedures for blinded and unblinded SSRE after any time period in a SW‐CRT are detailed. The performance of these procedures is then examined and contrasted using two example trial design scenarios. We find that if the two key variance parameters were underspecified by 50%, the SSRE procedures were able to increase power over the conventional SW‐CRT design by up to 41%, resulting in an empirical power above the desired level. Thus, though there are practical issues to consider, the performance of the procedures means researchers should consider incorporating SSRE in to future SW‐CRTs.     

设施葡萄萌芽调控中需冷量和需热量及其相互关系

采用3种需冷量估算模型（≤7．2℃模型、0～7．2℃模型和犹他模型）和2种需热量估算模型（生长度小时模型和有效积温模型）分别测定了14个设施葡萄品种的需冷量和需热量,并分析了两者的相互关系。结果表明,不同葡萄品种的需冷量和需热量存在一定差异,需冷量值介于754～1489h（≤7_2℃模型）、497-757h（0～7．2℃模型）或192～755．5CU（犹他模型）,需热量值为18491～24070GDH℃（生长度小时模型）或120～377D℃（有效积温模型）,且欧亚种品种需热量普遍高于欧美杂种品种。另外,本研究还表明,用生长度小时模型估算出来的需热量值与需冷量值之间呈负相关关系,而有效积温模型估算出来的需热量值与需冷量值之间则呈正相关关系。     

Genetic analysis of return over feed in Canadian Holsteins

The objective of this study was to investigate genetic merit of return over feed (ROF), which is a herd profit index defined by CanWest Dairy Herd Improvement as a difference between milk income and feed cost. A multiple-trait (MT) model and random regression model (RRM) were used. The traits analyzed in MT were rearing cost and ROF of the first three lactations. In RRM, a cumulative ROF was fitted as function of age and rearing cost was treated as a correlated trait. Variance components were estimated within a Bayesian framework by Gibbs sampling using a subsample of data. Breeding values were then estimated for 3 041 078 animals using records of 1 951 893 cows. Estimates of heritability for rearing cost from MT and RRM were 0.23 and 0.22, respectively. ROF per lactation and cumulative ROF were negatively correlated with rearing cost. Estimates of heritability of ROF through the first, second and third lactation from MT were 0.27, 0.10 and 0.08, respectively. Estimates of heritability of ROF from RRM increased with age and ranged from 0.08 through 0.31. Estimated breeding values (EBVs) for ROF from MT and RRM were moderately correlated with official EBV for production traits and the Canadian selection index (Lifetime Profit Index). Herd life EBV had -0.07 and 0.19 correlations with EBVs for ROF from MT and RRM, respectively. From both MT and RRM, small favorable correlations were reported between EBVs for ROF and for bone quality and angularity, whereas low unfavorable correlations were reported with EBV for udder depth, front end and chest width. Majority of correlations between EBVs for ROF and for reproduction traits were near 0, with the exception of EBV for gestation length, calf size and calving ease, where small favorable correlations were reported. The ROF is a good indicator of cow profitability despite the fact that it is a simplified profit index that does not account for animal-specific health and reproductive cost. However, because ROF does not account for differences in heritabilities between components of profit, ROF is not recommended to be used for direct selection for profit.     

脑磁图神经活动源数目的估计

在脑磁图信号的分析中,正确估计出脑磁图神经活动源的数目是进一步分析脑磁图信号的前提。目前广泛采用的信息论方法和主成分分析方法都是根据特征值来确定源的数目,这两种方法在源数目较多、噪声较强的情况下,会导致误判。该文提出了一种噪声调节自动阈值的脑磁图源数目判断方法,利用基于噪声调节的主成分分析并结合聂曼- 皮尔逊准则对脑磁图源数目进行估计。同时,该方法采用了基于小波的噪声方差估计,实现了脑磁图信号中噪声方差的精确估计。通过对基于信息论方法、主成分分析方法以及该文所提议方法的实验结果的比较,表明该文所提议方法能更准确地估计脑磁图源数目,特别是在源数目较多、信噪比较小的情况下,仍能准确地估计脑磁图源数目,具有较大的实际意义。