Search

Heritable genetic variation among free-living animals or plants is essential for populations to respond to selection and adapt. It is therefore important to be able to estimate additive genetic variance VA, which can be obtained using a generalized linear mixed model known as the animal model. An underlying assumption of the standard animal model is that the study population is genetically unstructured, which is often unrealistic. In fact, admixture might be the norm rather than the exception in the wild, like in geographically structured populations, in the presence of (im)migration, or in re-introduction and conservation contexts. Unfortunately, animal model estimators may be biased in such cases. So-called genetic group animal models that account for genetically differentiated subpopulations have recently become popular, but methodology is currently only available for cases where relatedness among individuals can be estimated from pedigrees.To ensure that genetic group animal models with heterogeneous VA remain applicable to populations with genomic data but no pedigrees, there is a clear need to generalize these models to the case when exclusively genomic data is available. We therefore introduce such methodology for wild admixed systems by extending methods that were recently suggested in the context of plant breeding. Our extension relaxes the limiting assumptions that currently restrict their use to artificial breeding setups.We illustrate the usefulness of the extended genomic genetic groups animal model on a wild admixed population of house sparrows resident in an island system in Northern Norway, where genome-wide data on more than 180 000 single nucleotide polymorphisms (SNPs) is available to derive genomic relatedness. We compare our estimates of quantitative genetic parameters to those derived from a corresponding pedigree-based genetic groups animal model. The satisfactory agreement indicates that the new method works as expected.Our extension of the very popular animal model ensures that the upcoming challenges with increasing availability of genomic data for quantitative genetic studies of wild admixed populations can be handled. To make the method widely available to the scientific community, we offer guidance in the form of a tutorial including step-by-step instructions to facilitate implementation.

In many biomedical studies or clinical trials, we have data with more than one response variable on the same subject repeatedly measured over time. In analyzing such data, we adopt a multivariate linear mixed-effects longitudinal model. On the other hand, in longitudinal data, we often find features that do not impact modeling the response variable and are eliminated from the study. In this paper, we consider the problem of simultaneous variable selection and estimation in a multivariate t linear mixed-effects model (MtLMM) for analyzing longitudinally measured multioutcome data. This work's motivation comes from a cohort study of patients with primary biliary cirrhosis. The interest is eliminating insignificant variables using the smoothly clipped and absolute deviation penalty function in the MtLMM. The proposed penalized model offers robustness and flexibility to accommodate fat tails. An expectation conditional maximization algorithm is employed for the computation of maximum likelihood estimates of parameters. The calculation of standard errors is affected by an information-based method. The methodology is illustrated by analyzing Mayo Clinic Primary Biliary Cirrhosis sequential (PBCseq) data and a simulation study. We found drugs and sex can be eliminated from the PBCseq analysis, and over time the disease progresses.

The stepped wedge cluster randomized trial (SW-CRT) is an increasingly popular design for evaluating health service delivery or policy interventions. An essential consideration of this design is the need to account for both within-period and between-period correlations in sample size calculations. Especially when embedded in health care delivery systems, many SW-CRTs may have subclusters nested in clusters, within which outcomes are collected longitudinally. However, existing sample size methods that account for between-period correlations have not allowed for multiple levels of clustering. We present computationally efficient sample size procedures that properly differentiate within-period and between-period intracluster correlation coefficients in SW-CRTs in the presence of subclusters. We introduce an extended block exchangeable correlation matrix to characterize the complex dependencies of outcomes within clusters. For Gaussian outcomes, we derive a closed-form sample size expression that depends on the correlation structure only through two eigenvalues of the extended block exchangeable correlation structure. For non-Gaussian outcomes, we present a generic sample size algorithm based on linearization and elucidate simplifications under canonical link functions. For example, we show that the approximate sample size formula under a logistic linear mixed model depends on three eigenvalues of the extended block exchangeable correlation matrix. We provide an extension to accommodate unequal cluster sizes and validate the proposed methods via simulations. Finally, we illustrate our methods in two real SW-CRTs with subclusters.

Any average pattern observed at the population level (cross-sectional analysis) may confound two different types of processes: some processes that occur among individuals and others that occur within individuals. Separating within- from among-individual processes is critical for our understanding of ecological and evolutionary dynamics. The within-individual centring method allows distinguishing within- from among-individual processes and this method has been widely used in ecology to investigate both linear and quadratic patterns. Here we show that two alternative equations could be used for the investigation of quadratic within-individual patterns. We explain the different assumptions and constraints of both equations. Reviewing the literature, we found that mainly one of these two equations has been used in studies investigating quadratic patterns. Yet this equation might not be the most appropriate in all circumstances leading to bias and imprecision. We show that these two alternative equations make different assumptions about the shape of the within-individual pattern. One equation assumes that the within-individual effect is related to an absolute process whereas the other assumes the effect arises from an individual relative process. The choice of using one equation instead of the other should depend upon the biological process investigated. Using simulations, we showed that a mismatch between the assumptions made by the equation used to analyse the data and the biological process investigated might led to flawed inference affecting output of model selection and accuracy of estimates. We stress that the equation used should be chosen carefully. We provide step by step guidelines for choosing an equation when studying quadratic pattern with the within-individual centring approach. We encourage the use of the within-individual centring method, promoting its relevant application for nonlinear relationships.

BACKGROUND Despite presenting higher risk of dementia, mild cognitive impairment (MCI) is not well defined in Down syndrome population. OBJECTIVE We aimed to describe cognitive and neuropsychological patterns associated with MCI in Down syndrome individuals. METHOD Two groups of adults with Down syndrome (control and prodromal) were studied throughout 3 years. Two linear mixed models and a model including the variables that best predicted group membership were built. RESULTS Behavioural Regulation Index (BRI) (Behaviour Rating Inventory of Executive Function test) and the model composed of BRI, abstraction and delayed verbal memory were the variable and model best predicting group membership, respectively. CONCLUSION Suggest a diagnosis of MCI when BRI is the earliest change perceived by caregivers and this is combined with low scores in abstract thinking, and when an amnesic pattern in delayed verbal memory is observed, but adaptive skills are preserved.

Consistent information on the current elemental composition of vegetation at global scale and the variables that determine it is lacking. To fill this gap, we gathered a total of 30 912 georeferenced records on woody plants foliar concentrations of nitrogen (N), phosphorus (P) and potassium (K) from published databases, and produced global maps of foliar N, P and K concentrations for woody plants using neural networks at a resolution of 1 km(2). We used data for climate, atmospheric deposition, soil and morphoclimatic groups to train the neural networks. Foliar N, P and K do not follow clear global latitudinal patterns but are consistent with the hypothesis of soil substrate age. We additionally built generalized linear mixed models to investigate the evolutionary history effect together with the effects of environmental effects. In this comparison, evolutionary history effects explained most of the variability in all cases (mostly > 60%). These results emphasize the determinant role of evolutionary history in foliar elemental composition, which should be incorporated in upcoming dynamic global vegetation models.

Meta-analysis is commonly used to compare two treatments. Network meta-analysis (NMA) is a powerful extension for comparing and contrasting multiple treatments simultaneously in a systematic review of multiple clinical trials. Although the practical utility of meta-analysis is apparent, it is not always straightforward to implement, especially for those interested in a Bayesian approach. This paper demonstrates that the recently-developed SAS procedure BGLIMM provides an intuitive and computationally efficient means for conducting Bayesian meta-analysis in SAS, using a worked example of a smoking cessation NMA data set. BGLIMM gives practitioners an effective and simple way to implement Bayesian meta-analysis (pairwise and network, either contrast-based or arm-based) without requiring significant background in coding or statistical modeling. Those familiar with generalized linear mixed models, and especially the SAS procedure GLIMMIX, will find this tutorial a useful introduction to Bayesian meta-analysis in SAS.

Mixed‐effects models for repeated measures (MMRMs) with an ‘unstructured’ (UN) covariance structure are frequently used in primary analyses for group comparisons of incomplete continuous longitudinal data from drug development trials. However, MMRM‐UN analysis could lead to convergence problems in numerical optimisation, especially in trials with a small sample size or high dropout rate. Although the so‐called sandwich covariance estimator is robust against the misspecification of the covariance structure, its performance deteriorates for small sample sizes. We review eight modified covariance estimators adjusted for small‐sample bias and compare their performances in the framework of MMRM analysis through simulations. In terms of the type 1 error rate and coverage probability of confidence intervals for group comparisons, Mancl and DeRouen's covariance estimator (MD) shows the best performance among the modified covariance estimators, followed by Fay and Graubard's estimator. The performance of MD is nearly equivalent to that of the Kenward–Roger method with a UN structure. The Kenward–Roger method with first‐order autoregressive structure results in substantial inflation of the type 1 error rate in the scenario where the variance of measurements increases across visits. In summary, we recommend the use of MD in MMRM analysis if the convergence problem involving a UN structure occurs in small clinical trials.

BACKGROUND Ordinances requiring the implementation of robbery prevention measures have been enacted at the city level in many jurisdictions. We evaluated the impact of an ordinance requiring crime prevention measures on subsequent crime rates. METHODS Crime reports for robbery and aggravated assault from January 2006 through December 2015 were linked to randomly-selected convenience stores and small retail grocers in Houston (n = 293). Store characteristics and compliance with a list of safety measures were collected by surveyors in 2011. Generalized linear mixed models were used to compare rates of crime before and after the implementation of the ordinance. RESULTS Robberies decreased significantly after the ordinance went into effect (rate ratio = 0.38; 95% confidence interval 0.29-0.51). No individual safety measure was associated with decreased robbery rates. No similar decrease was observed for aggravated assault. CONCLUSIONS City ordinances mandating crime prevention measures can be effective. We could not parse out the effectiveness of individual elements, suggesting a comprehensive approach may be more effective.

Rathouz and Gao [2] and Luo and Tsai [3] proposed valuable extensions to the generalized linear model for modeling a nonlinear monotonic relationship between the mean response and a set of covariates. In their extensions for discrete data the baseline response distribution is unspecified and is estimated from the data. We propose to extend this model for the analysis of longitudinal data by incorporating random effects into the linear predictor, and using maximum likelihood for estimation and inference. Motivated in particular by longitudinal studies of clinical scale outcomes, we developed an estimation procedure for a finite-support response using a generalized expectation-maximization algorithm where Gauss-Hermite quadrature is employed to approximate the integrals in the E step of the algorithm. Upon convergence, the observed information matrix is estimated through second-order numerical differentiation of the log-likelihood function. Asymptotic properties of the maximum likelihood estimates follow under the usual regularity conditions. Simulation studies are conducted to assess its finite-sample properties and compare the proposed model to the generalized linear mixed model. The proposed method is illustrated in an analysis of data from a longitudinal study of Huntington's disease.

Hidden Markov and semi-Markov models (H(S)MMs) constitute useful tools for modeling observations subject to certain dependency structures. The hidden states render these models very flexible and allow them to capture many different types of latent patterns and dynamics present in the data. This has led to the increased popularity of these models, which have been applied to a variety of problems in various domains and settings, including longitudinal data. In many longitudinal studies, the response variable is categorical or count-type. Generalized linear mixed models (GLMMs) can be used to analyze a wide range of variables, including categorical and count. The present study proposes a model that combines HSMMs with GLMMs, leading to generalized linear mixed hidden semi-Markov models (GLM-HSMMs). These models can account for time-varying unobserved heterogeneity and handle different response types. Parameter estimation is achieved using a Monte Carlo Newton-Raphson (MCNR)-like algorithm. In our proposed model, the distribution of the random effects depends on hidden states. We illustrate the applicability of GLM-HSMMs with an example in the field of occupational health, where the response variable consists of count values. Furthermore, we assess the performance of our MCNR-like algorithm through a simulation study.

A wide array of technologies are available for gaining insight into the movement of wild aquatic animals. Although acoustic telemetry can lack the fine‐scale spatial resolution of some satellite tracking technologies, the substantially longer battery life can yield important long‐term data on individual behavior and movement for low per‐unit cost. Typically, however, receiver arrays are designed to maximize spatial coverage at the cost of positional accuracy leading to potentially longer detection gaps as individuals move out of range between monitored locations. This is particularly true when these technologies are deployed to monitor species in hard‐to‐access locations.Here, we develop a novel approach to analyzing acoustic telemetry data, using the timing and duration of gaps between animal detections to infer different behaviors. Using the durations between detections at the same and different receiver locations (i.e., detection gaps), we classify behaviors into “restricted” or potential wider “out‐of‐range” movements synonymous with longer distance dispersal. We apply this method to investigate spatial and temporal segregation of inferred movement patterns in two sympatric species of reef shark within a large, remote, marine protected area (MPA). Response variables were generated using network analysis, and drivers of these movements were identified using generalized linear mixed models and multimodel inference.Species, diel period, and season were significant predictors of “out‐of‐range” movements. Silvertip sharks were overall more likely to undertake “out‐of‐range” movements, compared with gray reef sharks, indicating spatial segregation, and corroborating previous stable isotope work between these two species. High individual variability in “out‐of‐range” movements in both species was also identified.We present a novel gap analysis of telemetry data to help infer differential movement and space use patterns where acoustic coverage is imperfect and other tracking methods are impractical at scale. In remote locations, inference may be the best available tool and this approach shows that acoustic telemetry gap analysis can be used for comparative studies in fish ecology, or combined with other research techniques to better understand functional mechanisms driving behavior., Financial and logistical constraints may lead to imperfect in‐field acoustic receiver array designs which can result in acoustic telemetry data with long detection gaps. We present a novel gap analysis of telemetry data to help infer differential movement and space use patterns where acoustic coverage is imperfect and other tracking methods impractical

SummaryTo elucidate the molecular mechanisms underlying genetic variants identified from genome-wide association studies (GWAS) for a variety of phenotypic traits encompassing binary, continuous, count, and survival outcomes, we propose a novel and flexible method to test for mediation that can simultaneously accommodate multiple genetic variants and different types of outcome variables. Specifically, we employ the intersection-union test approach combined with likelihood ratio test to detect mediation effect of multiple genetic variants via some mediator (for example, the expression of a neighboring gene) on outcome. We fit high-dimensional generalized linear mixed models under the mediation framework, separately under the null and alternative hypothesis. We leverage Laplace approximation to compute the marginal likelihood of outcome and use coordinate descent algorithm to estimate corresponding parameters. Our extensive simulations demonstrate the validity of our proposed method and substantial, up to 97%, power gains over alternative methods. Applications to real data for the study of Chlamydia trachomatis infection further showcase advantages of our method. We believe our proposed method will be of value and general interest in this post-GWAS era to disentangle the potential causal mechanism from DNA to phenotype for new drug discovery and personalized medicine.

Premise With digitization and data sharing initiatives underway over the last 15 years, an important need has been prioritizing specimens to digitize. Because duplicate specimens are shared among herbaria in exchange and gift programs, we investigated the extent to which unique biogeographic data are held in small herbaria vs. these data being redundant with those held by larger institutions. We evaluated the unique specimen contributions that small herbaria make to biogeographic understanding at county, locality, and temporal scales. Methods We sampled herbarium specimens of 40 plant taxa from each of eight states of the United States of America in four broad status categories: extremely rare, very rare, common native, and introduced. We gathered geographic information from specimens held by large (≥100,000 specimens) and small (

1. In mixed stands, species complementarity (e.g. facilitation and competition reduction) may enhance forest tree productivity. Although positive mixture effects have been identified in forests world‐wide, the majority of studies have focused on two‐species interactions in managed systems with high functional diversity. We extended this line of research to examine mixture effects on tree productivity across landscape‐scale compositional and environmental gradients in the low functional diversity, fire‐suppressed, mixed‐conifer forests of the U.S. Interior West. 2. We investigated mixture effects on the productivity of Pinus ponderosa, Pseudotsuga menziesii and Abies concolor. Using region‐wide forest inventory data, we created individual‐tree generalized linear mixed models and examined the growth of these species across community gradients. We compared the relative influences of stand structure, age, competition and environmental stress on mixture effects using multi‐model inference. We analysed growth of neighbouring tree species to infer whether a mixture effect in a single species translated to the stand‐level. 3. We found support for a positive mixture effect in P. menziesii, although our results were equivocal in light of a weaker but still plausible alternative model. Growth of P. menziesii neighbouring species in mixed stands declined or held constant depending on aridity, suggesting that a positive mixture effect in P. menziesii does not necessarily extend to the stand level. We found no evidence for mixture effects in P. ponderosa, A. concolor or their neighbouring species. 4. Complementarity appears to have a limited influence on tree growth in the mixed‐conifer systems of the U.S. Interior West, reflecting limited functional diversity. Historical changes in stand structure following fire exclusion, particularly high stand densities, may limit the potential for positive species mixture effects. The limited species pool of Interior West forests increases the risk that, without careful management, what functional diversity exists could be lost to compositional changes resulting from stand dynamics or disturbance.

Change in breeding phenology is often a response to environmental forcing, but less is known of the mechanism underlying such changes and their fitness consequences. Here, we report on changes in the breeding phenology from a 27‐year longitudinal study (1991–2017) of individually marked, known‐aged grey seals (Halichoerus grypus) on Sable Island, Nova Scotia, Canada. We used generalized linear mixed models and a 3‐step process to develop a model that includes interactions between intrinsic and extrinsic covariates and to test hypotheses about the influence of fixed factors (maternal age, parity, previous reproductive success, pup sex, colony density, Atlantic Multidecal Oscillation (AMO), North Atlantic Oscillation (NAO), and Sea Surface Temperature) and a random factor (female identity) on parturition dates. We also examined the consequences of the shift in birthdates on maternal energy allocation in offspring as measured by pup weaning mass. Birthdates were known for 2,768 pups of 660 known‐age females. For 494 females with ≥2 parturition dates, repeatability as measured by the intraclass correlation was high (mean = 0.66). 87% of the variation in birthdates was explained by a mixed‐effects model that included intrinsic and extrinsic fixed effects. Most of the explained variation was associated with the random effect of female identity. Parity was the most important intrinsic fixed effect, with inexperienced mothers giving birth later in the season than multiparous females. Over almost 3 decades, mean birthdates advanced by 15 days. The mixed model with intrinsic effects and population size, the detrended AMO from the previous year and mean NAO in the previous 3 years explained 80% of the variation with 21% of variation from the fixed effects. Both primiparous and multiparous individuals responded to the climate forcing, and there was strong evidence for heterogeneity in the response. Nevertheless, the shift in birthdates did not impact pup weaning mass., Here, we studied how breeding phenology responds to maternal traits and demographic and environmental conditions over almost three decades and the resulting fitness consequences of those changes in a well‐studied population of grey seals. Mean birthdates advanced by 15 days. Longitudinal data on over 500 individual females showed that maternal birthdates were highly repeatable but despite that repeatability females responded to environmental forcing. Both primiparous and multiparous individuals responded to forcing by AMO, but there was strong evidence for heterogeneity in the response. Despite changing birthdates offspring weaning mass was not affected showing that climate driven changes in breeding phenology may not always have measurable consequences.

Aim The aim was to compare the performances of the World Health Organization (WHO) and population-based (PB) references in the screening for hydrocephalus in infants aged less than two years. Methods We collected 341 longitudinal head circumference (HC) measurements of hydrocephalic infants and 120,181 measurements of 15,145 healthy infants from primary care. The measurements were converted into z-scores, and a new screening parameter, change in HC standard deviation score (SDS) over time (ΔHC SDS), was calculated. Comparisons were made using receiver operating characteristics analysis and linear mixed models. Results The mean HC SDSWHO was 3.5 and the mean HC SDSPB was 2.9 in the hydrocephalic infants, and in healthy children those numbers were 1.0 SDSWHO and 0 SDSPB , respectively. The best screening accuracy was obtained with the PB reference in combination with the ΔHC SDS parameter (AUC 0.89). The accuracy of the WHO standard could be improved to a similar level by customizing the screening cut-offs of HC SDS according to the population and combining screening parameters. Conclusions Auxology alone was not sufficient for the screening of hydrocephalus. The WHO standard should be validated in the population, and population-specific cut-offs for normality defined before its introduction.

Urban Norway rats (Rattus norvegicus) carry several pathogens transmissible to people. However, pathogen prevalence can vary across fine spatial scales (i.e., by city block). Using a population genomics approach, we sought to describe rat movement patterns across an urban landscape and to evaluate whether these patterns align with pathogen distributions. We genotyped 605 rats from a single neighborhood in Vancouver, Canada, and used 1,495 genome‐wide single nucleotide polymorphisms to identify parent–offspring and sibling relationships using pedigree analysis. We resolved 1,246 pairs of relatives, of which only 1% of pairs were captured in different city blocks. Relatives were primarily caught within 33 meters of each other leading to a highly leptokurtic distribution of dispersal distances. Using binomial generalized linear mixed models, we evaluated whether family relationships influenced rat pathogen status with the bacterial pathogens Leptospira interrogans, Bartonella tribocorum, and Clostridium difficile, and found that an individual's pathogen status was not predicted any better by including disease status of related rats. The spatial clustering of related rats and their pathogens lends support to the hypothesis that spatially restricted movement promotes the heterogeneous patterns of pathogen prevalence evidenced in this population. Our findings also highlight the utility of evolutionary tools to understand movement and rat‐associated health risks in urban landscapes.

Sarcopenia is a geriatric syndrome characterized by significant loss of muscle mass. Based on a commonly used definition of the condition that involves three measurements, different subclinical and clinical states of sarcopenia are formed. These states constitute a partially ordered set (poset). This article focuses on the analysis of longitudinal poset in the context of sarcopenia. We propose an extension of the generalized linear mixed model and a recoding scheme for poset analysis such that two submodels-one for ordered categories and one for nominal categories-that include common random effects can be jointly estimated. The new poset model postulates random effects conceptualized as latent variables that represent an underlying construct of interest, that is, susceptibility to sarcopenia over time. We demonstrate how information can be gleaned from nominal sarcopenic states for strengthening statistical inference on a person's susceptibility to sarcopenia.

In cultivar testing, linear mixed models have been used routinely to analyze multienvironment trials. A single‐stage analysis is considered as the gold standard, whereas two‐stage analysis produces similar results when a fully efficient weighting method is used, namely when the full variance–covariance matrix of the estimated means from Stage 1 is forwarded to Stage 2. However, in practice, this may be hard to do and a diagonal approximation is often used. We conducted a cross‐validation with data from Swedish cultivar trials on winter wheat (Triticum aestivum L.) and spring barley (Hordeum vulgare L.) to assess the performance of single‐stage and two‐stage analyses. The fully efficient method and two diagonal approximation methods were used for weighting in the two‐stage analyses. In Sweden, cultivar recommendation is delineated by zones (regions), not individual locations. We demonstrate the use of best linear unbiased prediction (BLUP) for cultivar effects per zone, which exploits correlations between zones and thus allows information to be borrowed across zones. Complex variance–covariance structures were applied to allow for heterogeneity of cultivar × zone variance. The single‐stage analysis and the three weighted two‐stage analyses all performed similarly. Loss of information caused by a diagonal approximation of the variance–covariance matrix of adjusted means from Stage 1 was negligible. As expected, BLUP outperformed best linear unbiased estimation. Complex variance–covariance structures were dispensable. To our knowledge, this study is the first to use cross‐validation for comparing single‐stage analyses with stagewise analyses.

Stepped wedge cluster trials are an increasingly popular alternative to traditional parallel cluster randomized trials. Such trials often utilize a small number of clusters and numerous time intervals, and these components must be considered when choosing an analysis method. A generalized linear mixed model containing a random intercept and fixed time and intervention covariates is the most common analysis approach. However, the sole use of a random intercept applies a constant intraclass correlation coefficient structure, which is an assumption that is likely to be violated given stepped wedge trials (SWTs) have multiple time intervals. Alternatively, generalized estimating equations (GEE) are robust to the misspecification of the working correlation structure, although it has been shown that small-sample adjustments to standard error estimates and the use of appropriate degrees of freedom are required to maintain the validity of inference when the number of clusters is small. In this article, we show, using an extensive simulation study based on a motivating example and a more general design, the use of GEE can maintain the validity of inference in small-sample SWTs with binary outcomes. Furthermore, we show which combinations of bias corrections to standard error estimates and degrees of freedom work best in terms of attaining nominal type I error rates.

In order to evaluate and protect ecosystem services provided by peat and peaty soils, accurate estimations for the depth of the surface organic horizon are required. This study uses linear mixed models (LMMs) to test how topographic (elevation, slope, aspect) and superficial geology parameters can contribute to improved depth estimates across a Scottish upland catchment. Mean (n = 5) depth data from 283 sites (representing full covariate ranges) were used to calibrate LMMs, which were tested against a validation dataset. Models were estimated using maximum likelihood, and the Akaike Information Criterion was used to test whether the iterative addition of covariates to a model with constant fixed effects was beneficial. Elevation, slope and certain geology classes were all identified as useful covariates. Upon addition of the random effects (i.e. spatial modelling of residuals), the RMSE for the model with constant‐only fixed effects reduced by 24%. Addition of random effects to a model with topographic covariates (fixed effects = constant, slope, elevation) reduced the RMSE by 13%, whereas the addition of random effects to a model with topographic and geological covariates (fixed effects = constant, slope, elevation, certain geology classes) reduced the RMSE by only 3%. Therefore, much of the spatial pattern in depth was explained by the fixed effects in the latter model. The study contributes to a growing research base demonstrating that widely available topographic (and also here geological) datasets, which have national coverage, can be included in spatial models to improve organic horizon depth estimations.

Misspecification of the covariance structure in a linear mixed model (LMM) can lead to biased population parameters' estimates under MAR drop-out. In our motivating example of modeling CD4 cell counts during untreated HIV infection, random intercept and slope LMMs are frequently used. In this article, we evaluate the performance of LMMs with specific covariance structures, in terms of bias in the fixed effects estimates, under specific MAR drop-out mechanisms, and adopt a Bayesian model comparison criterion to discriminate between the examined approaches in real-data applications. We analytically show that using a random intercept and slope structure when the true one is more complex can lead to seriously biased estimates, with the degree of bias depending on the magnitude of the MAR drop-out. Under misspecified covariance structure, we compare in terms of induced bias the approach of adding a fractional Brownian motion (BM) process on top of random intercepts and slopes with the approach of using splines for the random effects. In general, the performance of both approaches was satisfactory, with the BM model leading to smaller bias in most cases. A simulation study is carried out to evaluate the performance of the proposed Bayesian criterion in identifying the model with the correct covariance structure. Overall, the proposed method performs better than the AIC and BIC criteria under our specific simulation setting. The models under consideration are applied to real data from the CASCADE study; the most plausible model is identified by all examined criteria.

Soil scientists are accustomed to geostatistical methods and tools such as semivariograms and kriging for analysis of observational data. Such methods assume and exploit that observations are spatially correlated. Conversely, analysis of variance (ANOVA) of designed experiments assumes that observations from different experimental units are independent, an assumption that is justified based on randomization. It may be beneficial, however, to perform an ANOVA assuming a geostatistical covariance model. Also, it is increasingly common to have multiple observations per experimental unit. Simple ANOVA assuming independence of observations is not appropriate for such data. Instead, a linear mixed model accounting for correlation among observations made on the same plot is required for proper analysis. The purpose of this paper is to demonstrate the benefits of integrating geostatistical covariance structures and ANOVA procedures into a linear mixed modelling framework. Two examples from designed experiments are considered in detail, making a link between terminologies and jargon used in geostatistical analysis on the one hand and linear mixed modelling on the other hand. We provide code in R and SAS for both examples in two supporting companion documents. HIGHLIGHTS: Analysis of variance and geostatistical analysis can be joined in a mixed model. Randomization justifies the independence assumption in analysis of variance. Geostatistical models imply a correlation of errors and can improve efficiency. Lacking randomization, spatial correlation can be accounted for in a mixed model.

Multivariate longitudinal data usually exhibit complex features such as the presence of censored responses due to detection limits of the assay and unavoidable missing values arising when participants make irregular visits that lead to intermittently recorded characteristics. A generalization of the multivariate linear mixed model constructed by taking into account impacts of censored and intermittent missing responses simultaneously, which is named as the MLMM-CM, has been recently proposed for more precisely analyzing such kinds of data. This paper aims at presenting a fully Bayesian sampling-based approach to the MLMM-CM for addressing the uncertainties of censored and missing responses as well as unknown parameters. Two widely accepted Bayesian computational techniques based on the Markov chain Monte Carlo and the inverse Bayes formulas coupled with the Gibbs (IBF-Gibbs) schemes are developed for carrying out posterior inference of the model. The proposed methodology is illustrated through a simulation study and a real-data example from the Adult AIDS Clinical Trials Group 388 study. Numerical results show empirically that the proposed Bayesian methodology performs satisfactorily and offers reliable posterior inference.

Species distribution models (SDMs) are used to test ecological theory and to direct targeted surveys for species of conservation concern. Several studies have tested for an influence of species traits on the predictive accuracy of SDMs. However, most used the same set of environmental predictors for all species and/or did not use truly independent data to test SDM accuracy. We built eight SDMs for each of 24 plant species of conservation concern, varying the environmental predictors included in each SDM version. We then measured the accuracy of each SDM using independent presence and absence data to calculate area under the receiver operating characteristic curve (AUC) and true positive rate (TPR). We used generalized linear mixed models to test for a relationship between species traits and SDM accuracy, while accounting for variation in SDM performance that might be introduced by different predictor sets. All traits affected one or both SDM accuracy measures. Species with lighter seeds, animal‐dispersed seeds, and a higher density of occurrences had higher AUC and TPR than other species, all else being equal. Long‐lived woody species had higher AUC than herbaceous species, but lower TPR. These results support the hypothesis that the strength of species–environment correlations is affected by characteristics of species or their geographic distributions. However, because each species has multiple traits, and because AUC and TPR can be affected differently, there is no straightforward way to determine a priori which species will yield useful SDMs based on their traits. Most species yielded at least one useful SDM. Therefore, it is worthwhile to build and test SDMs for the purpose of finding new populations of plant species of conservation concern, regardless of these species’ traits., We used independent field survey data to assess the accuracy of species distribution models (SDMs) built for 24 plant species of conservation concern. We tested whether traits of species (e.g., dispersal mechanism) or characteristics of their distribution (e.g., mean latitude) influenced model accuracy. Traits do influence species–environment relationships and therefore SDM accuracy, nevertheless a useful SDM could be built for most species.

Linear mixed models (LMMs) and their extensions have been widely used for high-dimensional genomic data analyses. While LMMs hold great promise for risk prediction research, the high dimensionality of the data and different effect sizes of genomic regions bring great analytical and computational challenges. In this work, we present a multikernel linear mixed model with adaptive lasso (KLMM-AL) to predict phenotypes using high-dimensional genomic data. We develop two algorithms for estimating parameters from our model and also establish the asymptotic properties of LMM with adaptive lasso when only one dependent observation is available. The proposed KLMM-AL can account for heterogeneous effect sizes from different genomic regions, capture both additive and nonadditive genetic effects, and adaptively and efficiently select predictive genomic regions and their corresponding effects. Through simulation studies, we demonstrate that KLMM-AL outperforms most of existing methods. Moreover, KLMM-AL achieves high sensitivity and specificity of selecting predictive genomic regions. KLMM-AL is further illustrated by an application to the sequencing dataset obtained from the Alzheimer's disease neuroimaging initiative.

In random-effects meta-analysis the between-study variance (τ 2) has a key role in assessing heterogeneity of study-level estimates and combining them to estimate an overall effect. For odds ratios the most common methods suffer from bias in estimating τ 2 and the overall effect and produce confidence intervals with below-nominal coverage. An improved approximation to the moments of Cochran's Q statistic, suggested by Kulinskaya and Dollinger (KD), yields new point and interval estimators of τ 2 and of the overall log-odds-ratio. Another, simpler approach (SSW) uses weights based only on study-level sample sizes to estimate the overall effect. In extensive simulations we compare our proposed estimators with established point and interval estimators for τ 2 and point and interval estimators for the overall log-odds-ratio (including the Hartung-Knapp-Sidik-Jonkman interval). Additional simulations included three estimators based on generalized linear mixed models and the Mantel-Haenszel fixed-effect estimator. Results of our simulations show that no single point estimator of τ 2 can be recommended exclusively, but Mandel-Paule and KD provide better choices for small and large numbers of studies, respectively. The KD estimator provides reliable coverage of τ 2. Inverse-variance-weighted estimators of the overall effect are substantially biased, as are the Mantel-Haenszel odds ratio and the estimators from the generalized linear mixed models. The SSW estimator of the overall effect and a related confidence interval provide reliable point and interval estimation of the overall log-odds-ratio.

This study develops and tests theory about different forms of absorptive capacity that support radical, differentiation and imitation product development strategies. Absorptive capacity theory provides a generalized explanation for how firms exploit their embeddedness in relationships with buyers and suppliers. We develop and test theory that relates combinations of four components of absorptive capacity (R&D capability, product development capability, cooperative embeddedness, and competitive embeddedness) to success rates in three product development strategies. We used data from the American pharmaceutical industry to estimate generalized linear mixed models. Our results confirm known relationships between R&D capability, alliance network position, and the development of radically new products, but reveal different sets of factors that influence differentiation and imitation. We describe a previously undetected influence of competitive embeddedness on the development of radically new products, a contrasting absorptive capacity structure for generic product development, and a mixed structure for differentiated product development.

Invasive alien species are a major threat to biodiversity and human activities, providing a strong incentive to understand the processes by which alien invasion occurs. While it is important to understand the determinants of success at each of several invasion stages—transport, introduction, establishment, and spread—few studies have explored the first of these stages. Here, we quantify and analyze variation in the success of individual animals in surviving the transport stage, based on shipping records of European passerines destined for New Zealand. We mined the original documents of Acclimatisation Societies, established in New Zealand for the purpose of introducing supposedly beneficial alien species, in combination with recently digitized newspaper archives, to produce a unique dataset of 122 ships that carried passerines from Europe to New Zealand between 1850 and 1885. For 37 of these shipments, data on the survival of individual species were available. Using generalized linear mixed models, we explored how survival was related to characteristics of the shipments and the species. We show that species differed greatly in their survival, but none of the tested traits accounted for these differences. Yet, survival increased over time, which mirrors the switch from early haphazard shipments to larger organized shipments. Our results imply that it was the quality of care received by the birds that most affected success at this stage of the invasion process., While it is important to understand the determinants of success at each of several invasion stages—transport, introduction, establishment, and spread—few studies have explored the first of these stages. Here, we quantify and analyse variation in the success of individual animals in surviving the transport stage, based on shipping records of European passerines destined for New Zealand. Our results imply that it was the quality of care received by the birds that most affected success at this stage of the invasion process.

Hierarchical models are recommended for meta-analyzing diagnostic test accuracy (DTA) studies. The bivariate random-effects model is currently widely used to synthesize a pair of test sensitivity and specificity using logit transformation across studies. This model assumes a bivariate normal distribution for the random-effects. However, this assumption is restrictive and can be violated. When the assumption fails, inferences could be misleading. In this paper, we extended the current bivariate random-effects model by assuming a flexible bivariate skew-normal distribution for the random-effects in order to robustly model logit sensitivities and logit specificities. The marginal distribution of the proposed model is analytically derived so that parameter estimation can be performed using standard likelihood methods. The method of weighted-average is adopted to estimate the overall logit-transformed sensitivity and specificity. An extensive simulation study is carried out to investigate the performance of the proposed model compared to other standard models. Overall, the proposed model performs better in terms of confidence interval width of the average logit-transformed sensitivity and specificity compared to the standard bivariate linear mixed model and bivariate generalized linear mixed model. Simulations have also shown that the proposed model performed better than the well-established bivariate linear mixed model in terms of bias and comparable with regards to the root mean squared error (RMSE) of the between-study (co)variances. The proposed method is also illustrated using a published meta-analysis data.

We derive precise asymptotic results that are directly usable for confidence intervals and Wald hypothesis tests for likelihood-based generalized linear mixed model analysis. The essence of our approach is to derive the exact leading term behaviour of the Fisher information matrix when both the number of groups and number of observations within each group diverge. This leads to asymptotic normality results with simple studentizable forms. Similar analyses result in tractable leading term forms for the determination of approximate locally D-optimal designs.

Due to reductions in both time and cost, group testing is a popular alternative to individual-level testing for disease screening. These reductions are obtained by testing pooled biospecimens (e.g., blood, urine, swabs, etc.) for the presence of an infectious agent. However, these reductions come at the expense of data complexity, making the task of conducting disease surveillance more tenuous when compared to using individual-level data. This is because an individual’s disease status may be obscured by a group testing protocol and the effect of imperfect testing. Furthermore, unlike individual-level testing, a given participant could be involved in multiple testing outcomes and/or may never be tested individually. To circumvent these complexities and to incorporate all available information, we propose a Bayesian generalized linear mixed model that accommodates data arising from any group testing protocol, estimates unknown assay accuracy probabilities, and accounts for potential heterogeneity in the covariate effects across population subgroups (e.g., clinic sites, etc.); this latter feature is of key interest to practitioners tasked with conducting disease surveillance. To achieve model selection, our proposal uses spike and slab priors for both fixed and random effects. The methodology is illustrated through numerical studies and is applied to chlamydia surveillance data collected in Iowa.

Odonates are an ancient group of freshwater invertebrates that have complex life cycles and highly variable phenotypes and behaviours. These traits make odonates a very attractive model group to study various aspects and phenomena in the fields of ecology, behaviour, and evolutionary biology. Such variability in appearance and behaviour can influence the detectability of individual species under natural conditions; and this variability in detection rates, if not statistically treated, may lead to significant distortions of results and misleading conclusions. In the present study, we used detectability models to predict the time to detection (TTD) of individual odonate species and generalised linear mixed models to analyse the effects of morphological and behavioural traits on the TTD of the first individual of a species. We found that only abundance had a significant effect on the TTD of both Anisoptera and Zygoptera, while the larger Anisoptera were generally detected earlier. No such patterns were observed in Zygoptera; however, dispersal ability was found to have had some influence. Surprisingly, coloration and activity were not found to influence the TTD of odonates. Therefore, although there are considerable differences between species, with the exception of abundance and size, we did not detect any easily measurable traits that would unequivocally explain the differences in the detection rates of individual species.

Multienvironment trials (METs) are used to investigate the performance of crop genotypes. To efficiently generate reliable performance estimates, the magnitude and patterns of genotype × environment interaction (G×E) in MET data must be known. We quantified G×E in fall-planted common wheat (Triticum aestivum L.) in California, with the goal of increasing the reliability and efficiency of statewide variety testing activities. Linear mixed models and the genotype main effects plus G×E interaction effects (GGE) biplot method were used to analyze MET data for 211 common wheat genotypes, the MET consisted of 9 locations and 14 seasons. The representativeness and discriminating power of the MET locations were tested, and estimates of the optimum number of test locations were made. The analyses did not find evidence for significant, repeatable, crossover G×E. The genotype and G×E effects were of a similar magnitude, and the G×E effects were relatively strong compared with other sources of variance but were dominated by seasonal effects, with potentially repeatable genotype-by-location (G×L) effects being relatively weak. The GGE analyses did not detect repeatable G×L patterns across seasons. As a result, we conclude that the cereal production regions of California consist of a single, but unstable, mega-environment for common wheat grain yield. The test location evaluation found few significant differences between test locations in terms of how well they represent the target production environment. We estimate that the number of test locations could be reduced while maintaining trial accuracy, which would improve the resource use efficiency of statewide trial activities without sacrificing information about variety-specific common wheat yield performance.

The principal goals of a plant breeding program are to provide breeders with cultivar information for selection purposes and to provide farmers with high-yielding and stable cultivars. For that reason, multi-environment trials need to be done to predict future cultivar yield, and a robust statistical procedure is needed to provide reliable information on the tested cultivars. In Sweden, the statistical procedure follows the tradition of modeling cultivar effects as fixed. Moreover, the analysis is performed separately by zone and level of fungicide treatment, and so the factorial information regarding cultivar x zone x fungicide combinations is not explored. Thus, the question arose whether the statistical method could be improved to increase accuracy in zone-based cultivar prediction, since the cultivar recommendation is zone based. In this paper, the performance of empirical best linear unbiased estimation (E-BLUE) and empirical best linear unbiased prediction (E-BLUP) are compared using cross-validation for winter wheat (Triticum aestivum L.) and spring barley (Hordeum vulgare L.), in single-year and multiyear series of trials. Data were obtained from three agricultural zones of Sweden. Several linear mixed models were compared, and model performance was evaluated using the mean squared error of prediction criterion. The E-BLUP method outperformed the E-BLUE method in both crops and series. The prediction accuracy for zone-based yield was improved by using E-BLUP because the random-effects assumption for cultivar x zone interaction allows information to be borrowed across zones. We conclude that E-BLUP should replace the currently used E-BLUE approach to predict zone-based cultivar yield.