Results 1  10
of
197
Random effects structure for confirmatory hypothesis testing: Keep it maximal.
 Journal of Memory and Language,
, 2013
"... Abstract Linear mixedeffects models (LMEMs) have become increasingly prominent in psycholinguistics and related areas. However, there is currently little understanding of how different random effects structures affect generalizability. Here, we argue that researchers using LMEMs for confirmatory h ..."
Abstract

Cited by 151 (5 self)
 Add to MetaCart
(Show Context)
Abstract Linear mixedeffects models (LMEMs) have become increasingly prominent in psycholinguistics and related areas. However, there is currently little understanding of how different random effects structures affect generalizability. Here, we argue that researchers using LMEMs for confirmatory hypothesis testing should minimally adhere to the standards that have been in place for many decades. Through theoretical arguments and Monte Carlo simulation, we show that LMEMs generalize best when they include the maximal random effects structure justified by the design. In contrast, LMEMs including the maximal random
MCMC Methods for Multiresponse Generalized Linear Mixed Models: The MCMCglmm R Package
"... Generalized linear mixed models provide a flexible framework for modeling a range of data, although with nonGaussian response variables the likelihood cannot be obtained in closed form. Markov chain Monte Carlo methods solve this problem by sampling from a series of simpler conditional distribution ..."
Abstract

Cited by 83 (0 self)
 Add to MetaCart
(Show Context)
Generalized linear mixed models provide a flexible framework for modeling a range of data, although with nonGaussian response variables the likelihood cannot be obtained in closed form. Markov chain Monte Carlo methods solve this problem by sampling from a series of simpler conditional distributions that can be evaluated. The R package MCMCglmm, implements such an algorithm for a range of model fitting problems. More than one response variable can be analysed simultaneously, and these variables are allowed to follow Gaussian, Poisson, multi(bi)nominal, exponential, zeroinflated and censored distributions. A range of variance structures are permitted for the random effects, including interactions with categorical or continuous variables (i.e., random regression), and more complicated variance structures that arise through shared ancestry, either through a pedigree or through a phylogeny. Missing values are permitted in the response variable(s) and data can be known up to some level of measurement error as in metaanalysis. All simulation is done in C / C++ using the CSparse library for sparse linear systems. If you use the software please cite this article, as published in the Journal of Statistic Software
The nouturn sampler: Adaptively setting path lengths in Hamiltonian Monte Carlo. Arxiv preprint arXiv:1111.4246
, 2011
"... Hierarchical Bayesian models are a mainstay of the machine learning and statistics communities. Exact posterior inference in such models is rarely tractable, so researchers and practitioners must usually resort to approximate inference methods. Perhaps the most popular class of approximate posterior ..."
Abstract

Cited by 70 (2 self)
 Add to MetaCart
(Show Context)
Hierarchical Bayesian models are a mainstay of the machine learning and statistics communities. Exact posterior inference in such models is rarely tractable, so researchers and practitioners must usually resort to approximate inference methods. Perhaps the most popular class of approximate posterior inference algorithms, Markov Chain Monte Carlo (MCMC) methods offer schemes for
Price and Authority
 in InterProfit Center Transactions.” American Journal of Sociology 94: S17S51
, 1988
"... Number of pages: 35 Number of tables: 1 Number of figures: 10 Number of references: 40 ..."
Abstract

Cited by 13 (0 self)
 Add to MetaCart
(Show Context)
Number of pages: 35 Number of tables: 1 Number of figures: 10 Number of references: 40
Bayesian Estimation Supersedes the t Test
"... This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Bayesian estimation for 2 groups provides complete distributions of credible valu ..."
Abstract

Cited by 12 (2 self)
 Add to MetaCart
(Show Context)
This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Bayesian estimation for 2 groups provides complete distributions of credible values for the effect size, group means and their difference, standard deviations and their difference, and the normality of the data. The method handles outliers. The decision rule can accept the null value (unlike traditional t tests) when certainty in the estimate is high (unlike Bayesian model comparison using Bayes factors). The method also yields precise estimates of statistical power for various research goals. The software and programs are free and run on Macintosh, Windows, and Linux platforms.
Efficient Ranking in Sponsored Search
"... Abstract. In the standard model of sponsored search auctions, an ad is ranked according to the product of its bid and its estimated clickthrough rate (known as the quality score), where the estimates are taken as exact. This paper reexamines the form of the efficient ranking rule when uncertainty ..."
Abstract

Cited by 12 (1 self)
 Add to MetaCart
(Show Context)
Abstract. In the standard model of sponsored search auctions, an ad is ranked according to the product of its bid and its estimated clickthrough rate (known as the quality score), where the estimates are taken as exact. This paper reexamines the form of the efficient ranking rule when uncertainty in clickthrough rates is taken into account. We provide a sufficient condition under which applying an exponent—strictly less than one—to the quality score improves expected efficiency. The condition holds for a large class of distributions known as natural exponential families, and for the lognormal distribution. An empirical analysis of Yahoo’s sponsored search logs reveals that exponent settings substantially smaller than one can be efficient for both high and low volume keywords, implying substantial deviations from the traditional ranking rule. 1
Quantifying Inter and IntraPopulation Niche Variability Using Hierarchical Bayesian Stable Isotope Mixing Models
, 2009
"... Variability in resource use defines the width of a trophic niche occupied by a population. Intrapopulation variability in resource use may occur across hierarchical levels of population structure from individuals to subpopulations. Understanding how levels of population organization contribute to p ..."
Abstract

Cited by 8 (1 self)
 Add to MetaCart
(Show Context)
Variability in resource use defines the width of a trophic niche occupied by a population. Intrapopulation variability in resource use may occur across hierarchical levels of population structure from individuals to subpopulations. Understanding how levels of population organization contribute to population niche width is critical to ecology and evolution. Here we describe a hierarchical stable isotope mixing model that can simultaneously estimate both the prey composition of a consumer diet and the diet variability among individuals and across levels of population organization. By explicitly estimating variance components for multiple scales, the model can deconstruct the niche width of a consumer population into relevant levels of population structure. We apply this new approach to stable isotope data from a population of gray wolves from coastal British Columbia, and show support for extensive intrapopulation niche variability among individuals, social groups, and geographically isolated subpopulations. The analytic method we describe improves mixing models by accounting for diet variability, and improves isotope niche width analysis by quantitatively assessing the contribution of levels of organization to the niche width of a population.
Effects of competition on tree radialgrowth vary in importance but not in intensity along climatic gradients.
 J. Ecol.
, 2011
"... Summary 1. Plant interactions play a central role in regulating plant communities and this role can be altered by abiotic stress. With increasing stress, ecological theory predicts that the role of competition decreases whilst that of facilitation increases. Such predictions have been tested with s ..."
Abstract

Cited by 7 (1 self)
 Add to MetaCart
(Show Context)
Summary 1. Plant interactions play a central role in regulating plant communities and this role can be altered by abiotic stress. With increasing stress, ecological theory predicts that the role of competition decreases whilst that of facilitation increases. Such predictions have been tested with shortterm plant removal experiments using two distinct indices evaluating the role of plant interactions: the intensity (absolute impact) and the importance (impact relative to that of other abiotic constraints) of plant interactions. 2. Using data on individual tree radial growth from more than 17 000 forest plots covering the habitat conditions of 16 species in the Alps and the Jura mountains of France, we show that nonmanipulative estimates of plant interactions provide an alternative to this experimental approach. We developed a Bayesian neighbourhood growth competition model to test theoretical predictions about plantplant interactions with a much larger spatiotemporal scope and set of study species than classically used in experimental studies of plantplant interactions. 3. Our analyses revealed that competition measured as neighbours effects on adult tree growthvaries in importance but not in intensity along two major bioclimatic gradients (degreeday sum and water availability). Observed patterns of competition importance differed between shadetolerant and shadeintolerant tree species. First, the mean importance of competition was found to be much higher for shadeintolerant species. Second, for shadeintolerant species the importance of competition remained high even at low crowding indices (i.e. at a low competitor density), whereas for shadetolerant species competition only became important at high crowding indices. 4. Synthesis. Our nonmanipulative approach to the study of plantplant interactions allows analysing interactions among many species over large climatic gradients. Our results clearly demonstrate that a quantitative estimation of density dependence effects is key to understanding how plantplant interactions vary along abiotic gradients. Growth predictions derived from our model can easily be integrated with other results on tree regeneration and mortality in individualbased models to investigate how plantplant interactions drive tree population and community dynamics under varying climatic conditions.
Pärt T. Effect of geolocators on migration and subsequent breeding performance of a longdistance passerine migrant
 PLoS ONE. 2013; 8: e82316. doi: 10.1371/journal.pone.0082316 PMID: 24324770
"... Geolocators are small lightweight data loggers used to track individual migratory routes, and their use has increased exponentially in birds. However, the effects of geolocators on individual performance are still poorly known. We studied geolocator effects on a longdistance migrating passerine bi ..."
Abstract

Cited by 7 (0 self)
 Add to MetaCart
(Show Context)
Geolocators are small lightweight data loggers used to track individual migratory routes, and their use has increased exponentially in birds. However, the effects of geolocators on individual performance are still poorly known. We studied geolocator effects on a longdistance migrating passerine bird, the northern wheatear (Oenanthe oenanthe L.). We asked the general question of whether geolocators affect migratory behaviour and subsequent reproductive performance of small passerines by comparing arrival time, breeding time, breeding success and survival of geolocator versus control birds of known identity and breeding history. During two years geolocator birds (n=37) displayed a lower apparent survival (30%) as compared to controls (45%, n=164). Furthermore, returning geolocator birds (n=12) arrived on average 3.5 days later, started laying eggs 6.3 days later, and had lower nest success (25%) than control birds (78%). Our results suggest that geolocators affect migratory performance with carryover effects to the timing of breeding and reproductive success in the subsequent breeding season. We discuss the implications of such geolocator effects for the study of migratory strategies of small passerines in general and suggest how to identify and investigate such effects in the future.
Shoot, shovel and shut up: cryptic poaching slows restoration of a large carnivore in Europe. Proc Biol Sci 279: 910–915
, 2012
"... Poaching is a widespread and wellappreciated problem for the conservation of many threatened species. Because poaching is illegal, there is strong incentive for poachers to conceal their activities, and consequently, little data on the effects of poaching on population dynamics are available. Quan ..."
Abstract

Cited by 6 (0 self)
 Add to MetaCart
Poaching is a widespread and wellappreciated problem for the conservation of many threatened species. Because poaching is illegal, there is strong incentive for poachers to conceal their activities, and consequently, little data on the effects of poaching on population dynamics are available. Quantifying poaching mortality should be a required knowledge when developing conservation plans for endangered species but is hampered by methodological challenges. We show that rigorous estimates of the effects of poaching relative to other sources of mortality can be obtained with a hierarchical statespace model combined with multiple sources of data. Using the Scandinavian wolf (Canis lupus) population as an illustrative example, we show that poaching accounted for approximately half of total mortality and more than twothirds of total poaching remained undetected by conventional methods, a source of mortality we term as 'cryptic poaching'. Our simulations suggest that without poaching during the past decade, the population would have been almost four times as large in 2009. Such a severe impact of poaching on population recovery may be widespread among large carnivores. We believe that conservation strategies for large carnivores considering only observed data may not be adequate and should be revised by including and quantifying cryptic poaching.