Genetic improvement of feed conversion ratio via indirect selection against lipid deposition in farmed rainbow trout (Oncorhynchus mykiss Walbaum)

The research leading to these results has received funding from the European Union's Seventh Framework Programme (KBBE.2013.1.2-10) under grant agreement n° 613611 FISHBOOST. Moreover, the original data collection was supported by the European Union, Project PROGRESS Q5RS-2001-00994. The staff at Tervo station, Ossi Ritola and Tuija Paananen, are highly acknowledged for fish management. A. Ka., A. Ki., S. M., D. H. and K. R. designed research and wrote the paper; A.Ka analyzed the data and had primary responsibility for the final content. All authors have read and approved the manuscript. The authors declare no conflicts of interest.

Strong inference from transect sign surveys : combining spatial autocorrelation and misclassification occupancy models to quantify the detectability of a recovering carnivore

Funding — Forest Enterprise Scotland and the University of Aberdeen provided funding for the project. The Carnegie Trust supported the lead author, E. McHenry, in this research through the award of a tuition fees bursary.

Bayesian Inference in Large-scale Problems

Many modern applications fall into the category of "large-scale" statistical problems, in which both the number of observations n and the number of features or parameters p may be large. Many existing methods focus on point estimation, despite the continued relevance of uncertainty quantification in the sciences, where the number of parameters to estimate often exceeds the sample size, despite huge increases in the value of n typically seen in many fields. Thus, the tendency in some areas of industry to dispense with traditional statistical analysis on the basis that "n=all" is of little relevance outside of certain narrow applications. The main result of the Big Data revolution in most fields has instead been to make computation much harder without reducing the importance of uncertainty quantification. Bayesian methods excel at uncertainty quantification, but often scale poorly relative to alternatives. This conflict between the statistical advantages of Bayesian procedures and their substantial computational disadvantages is perhaps the greatest challenge facing modern Bayesian statistics, and is the primary motivation for the work presented here.

Two general strategies for scaling Bayesian inference are considered. The first is the development of methods that lend themselves to faster computation, and the second is design and characterization of computational algorithms that scale better in n or p. In the first instance, the focus is on joint inference outside of the standard problem of multivariate continuous data that has been a major focus of previous theoretical work in this area. In the second area, we pursue strategies for improving the speed of Markov chain Monte Carlo algorithms, and characterizing their performance in large-scale settings. Throughout, the focus is on rigorous theoretical evaluation combined with empirical demonstrations of performance and concordance with the theory.

One topic we consider is modeling the joint distribution of multivariate categorical data, often summarized in a contingency table. Contingency table analysis routinely relies on log-linear models, with latent structure analysis providing a common alternative. Latent structure models lead to a reduced rank tensor factorization of the probability mass function for multivariate categorical data, while log-linear models achieve dimensionality reduction through sparsity. Little is known about the relationship between these notions of dimensionality reduction in the two paradigms. In Chapter 2, we derive several results relating the support of a log-linear model to nonnegative ranks of the associated probability tensor. Motivated by these findings, we propose a new collapsed Tucker class of tensor decompositions, which bridge existing PARAFAC and Tucker decompositions, providing a more flexible framework for parsimoniously characterizing multivariate categorical data. Taking a Bayesian approach to inference, we illustrate empirical advantages of the new decompositions.

Latent class models for the joint distribution of multivariate categorical, such as the PARAFAC decomposition, data play an important role in the analysis of population structure. In this context, the number of latent classes is interpreted as the number of genetically distinct subpopulations of an organism, an important factor in the analysis of evolutionary processes and conservation status. Existing methods focus on point estimates of the number of subpopulations, and lack robust uncertainty quantification. Moreover, whether the number of latent classes in these models is even an identified parameter is an open question. In Chapter 3, we show that when the model is properly specified, the correct number of subpopulations can be recovered almost surely. We then propose an alternative method for estimating the number of latent subpopulations that provides good quantification of uncertainty, and provide a simple procedure for verifying that the proposed method is consistent for the number of subpopulations. The performance of the model in estimating the number of subpopulations and other common population structure inference problems is assessed in simulations and a real data application.

In contingency table analysis, sparse data is frequently encountered for even modest numbers of variables, resulting in non-existence of maximum likelihood estimates. A common solution is to obtain regularized estimates of the parameters of a log-linear model. Bayesian methods provide a coherent approach to regularization, but are often computationally intensive. Conjugate priors ease computational demands, but the conjugate Diaconis--Ylvisaker priors for the parameters of log-linear models do not give rise to closed form credible regions, complicating posterior inference. In Chapter 4 we derive the optimal Gaussian approximation to the posterior for log-linear models with Diaconis--Ylvisaker priors, and provide convergence rate and finite-sample bounds for the Kullback-Leibler divergence between the exact posterior and the optimal Gaussian approximation. We demonstrate empirically in simulations and a real data application that the approximation is highly accurate, even in relatively small samples. The proposed approximation provides a computationally scalable and principled approach to regularized estimation and approximate Bayesian inference for log-linear models.

Another challenging and somewhat non-standard joint modeling problem is inference on tail dependence in stochastic processes. In applications where extreme dependence is of interest, data are almost always time-indexed. Existing methods for inference and modeling in this setting often cluster extreme events or choose window sizes with the goal of preserving temporal information. In Chapter 5, we propose an alternative paradigm for inference on tail dependence in stochastic processes with arbitrary temporal dependence structure in the extremes, based on the idea that the information on strength of tail dependence and the temporal structure in this dependence are both encoded in waiting times between exceedances of high thresholds. We construct a class of time-indexed stochastic processes with tail dependence obtained by endowing the support points in de Haan's spectral representation of max-stable processes with velocities and lifetimes. We extend Smith's model to these max-stable velocity processes and obtain the distribution of waiting times between extreme events at multiple locations. Motivated by this result, a new definition of tail dependence is proposed that is a function of the distribution of waiting times between threshold exceedances, and an inferential framework is constructed for estimating the strength of extremal dependence and quantifying uncertainty in this paradigm. The method is applied to climatological, financial, and electrophysiology data.

The remainder of this thesis focuses on posterior computation by Markov chain Monte Carlo. The Markov Chain Monte Carlo method is the dominant paradigm for posterior computation in Bayesian analysis. It has long been common to control computation time by making approximations to the Markov transition kernel. Comparatively little attention has been paid to convergence and estimation error in these approximating Markov Chains. In Chapter 6, we propose a framework for assessing when to use approximations in MCMC algorithms, and how much error in the transition kernel should be tolerated to obtain optimal estimation performance with respect to a specified loss function and computational budget. The results require only ergodicity of the exact kernel and control of the kernel approximation accuracy. The theoretical framework is applied to approximations based on random subsets of data, low-rank approximations of Gaussian processes, and a novel approximating Markov chain for discrete mixture models.

Data augmentation Gibbs samplers are arguably the most popular class of algorithm for approximately sampling from the posterior distribution for the parameters of generalized linear models. The truncated Normal and Polya-Gamma data augmentation samplers are standard examples for probit and logit links, respectively. Motivated by an important problem in quantitative advertising, in Chapter 7 we consider the application of these algorithms to modeling rare events. We show that when the sample size is large but the observed number of successes is small, these data augmentation samplers mix very slowly, with a spectral gap that converges to zero at a rate at least proportional to the reciprocal of the square root of the sample size up to a log factor. In simulation studies, moderate sample sizes result in high autocorrelations and small effective sample sizes. Similar empirical results are observed for related data augmentation samplers for multinomial logit and probit models. When applied to a real quantitative advertising dataset, the data augmentation samplers mix very poorly. Conversely, Hamiltonian Monte Carlo and a type of independence chain Metropolis algorithm show good mixing on the same dataset.

Bayesian Emulation for Sequential Modeling, Inference and Decision Analysis

The advances in three related areas of state-space modeling, sequential Bayesian learning, and decision analysis are addressed, with the statistical challenges of scalability and associated dynamic sparsity. The key theme that ties the three areas is Bayesian model emulation: solving challenging analysis/computational problems using creative model emulators. This idea defines theoretical and applied advances in non-linear, non-Gaussian state-space modeling, dynamic sparsity, decision analysis and statistical computation, across linked contexts of multivariate time series and dynamic networks studies. Examples and applications in financial time series and portfolio analysis, macroeconomics and internet studies from computational advertising demonstrate the utility of the core methodological innovations.

Chapter 1 summarizes the three areas/problems and the key idea of emulating in those areas. Chapter 2 discusses the sequential analysis of latent threshold models with use of emulating models that allows for analytical filtering to enhance the efficiency of posterior sampling. Chapter 3 examines the emulator model in decision analysis, or the synthetic model, that is equivalent to the loss function in the original minimization problem, and shows its performance in the context of sequential portfolio optimization. Chapter 4 describes the method for modeling the steaming data of counts observed on a large network that relies on emulating the whole, dependent network model by independent, conjugate sub-models customized to each set of flow. Chapter 5 reviews those advances and makes the concluding remarks.

Testing for biases in selection on avian reproductive traits and partitioning direct and indirect selection using quantitative genetic models

Key life history traits such as breeding time and clutch size are frequently both heritable and under directional selection, yet many studies fail to document micro-evolutionary responses. One general explanation is that selection estimates are biased by the omission of correlated traits that have causal effects on fitness, but few valid tests of this exist. Here we show, using a quantitative genetic framework and six decades of life-history data on two free-living populations of great tits Parus major, that selection estimates for egg-laying date and clutch size are relatively unbiased. Predicted responses to selection based on the Robertson-Price Identity were similar to those based on the multivariate breeder’s equation, indicating that unmeasured covarying traits were not missing from the analysis. Changing patterns of phenotypic selection on these traits (for laying date, linked to climate change) therefore reflect changing selection on breeding values, and genetic constraints appear not to limit their independent evolution. Quantitative genetic analysis of correlational data from pedigreed populations can be a valuable complement to experimental approaches to help identify whether apparent associations between traits and fitness are biased by missing traits, and to parse the roles of direct versus indirect selection across a range of environments.

Non-Consumptive Effects of Predators in Coral Reef Communities and the Indirect Consequences of Marine Protected Areas

Predators exert strong direct and indirect effects on ecological communities by intimidating their prey. Non-consumptive effects (NCEs) of predators are important features of many ecosystems and have changed the way we understand predator-prey interactions, but are not well understood in some systems. For my dissertation research I combined a variety of approaches to examine the effect of predation risk on herbivore foraging and reproductive behaviors in a coral reef ecosystem. In the first part of my dissertation, I investigated how diet and territoriality of herbivorous fish varied across multiple reefs with different levels of predator biomass in the Florida Keys National Marine Sanctuary. I show that both predator and damselfish abundance impacted diet diversity within populations for two herbivores in different ways. Additionally, reef protection and the associated recovery of large predators appeared to shape the trade-off reef herbivores made between territory size and quality. In the second part of my dissertation, I investigated context-dependent causal linkages between predation risk, herbivore foraging behavior and resource consumption in multiple field experiments. I found that reef complexity, predator hunting mode, light availability and prey hunger influenced prey perception of threat and their willingness to feed. This research argues for more emphasis on the role of predation risk in affecting individual herbivore foraging behavior in order to understand the implications of human-mediated predator removal and recovery in coral reef ecosystems.

A Recursive Income Model for Canadians : The Direct and Indirect Effects of Family Background (Working Paper 28)

This paper presents the results of a multi-equation income model which has been estimated for Canadian men and women which incorporates the effects of a number of important family background variables, including mother’s and father’s education, parents’ immigration status, their age at immigration, place of birth, language development, and learning background. Not only education, but also the individual’s tested literacy and numeracy levels are treated as intermediate outcomes which are affected by background and which, in turn, affect income. Many of the background variables are found to have important indirect effects on income which would be missed by more conventional approaches. We also find some interesting gender aspects with respect to the influences of parents’ educations on their children’s outcomes. Various policy implications of the findings are discussed.

The New Architecture of EU Equality Law after CHEZ: Did the Court of Justice reconceptualise direct and indirect discrimination?

The Court of Justice’s decision of the 16 July 2015, in Case C-83/14 CHEZ Razpredelenie Bulgaria AD v Komisia za zashtita ot diskriminatsia, is a critically important case for two main reasons. First, it represents a further step along the path of addressing ethnic discrimination against Roma communities in Europe, particularly in Bulgaria, where the case arises. Second, it provides interpretations (sometimes controversial interpretations) of core concepts in the EU antidiscrimination Directives that will be drawn on in the application of equality law well beyond Bulgaria, and well beyond the pressing problem of ethnic discrimination against Roma. This article focuses particularly on the second issue, the potentially broader implications of the case. In particular, it will ask whether the Court of Justice’s approach in CHEZ is subtly redrawing the boundaries of EU equality law in general, in particular by expanding the concept of direct discrimination, or whether the result and the approach adopted is sui generis, one depending on the particular context of the case and the fact that it involves allegations of discrimination against Roma, and therefore of limited general application.

Shame, Dissociation, and Complex PTSD Symptoms in Traumatized Psychiatric and Control Groups: Direct and Indirect Associations With Relationship Distress

Objectives: Elevated shame and dissociation are common in dissociative identity disorder (DID) and chronic posttraumatic stress disorder (PTSD) and are part of the constellation of symptoms defined as complex PTSD. Previous work examined the relationship between shame, dissociation, and complex PTSD and whether they are associated with intimate relationship anxiety, relationship depression, and fear of relationships. This study investigated these variables in traumatized clinical samples and a nonclinical community group.

Method: Participants were drawn from the DID (n = 20), conflict-related chronic PTSD (n = 65), and nonclinical (n = 125) populations and completed questionnaires assessing the variables of interest. A model examining the direct impact of shame and dissociation on relationship functioning, and their indirect effect via complex PTSD symptoms, was tested through path analysis.

Results: The DID sample reported significantly higher dissociation, shame, complex PTSD symptom severity, relationship anxiety, relationship depression, and fear of relationships than the other two samples. Support was found for the proposed model, with shame directly affecting relationship anxiety and fear of relationships, and pathological dissociation directly affecting relationship anxiety and relationship depression. The indirect effect of shame and dissociation via complex PTSD symptom severity was evident on all relationship variables.

Conclusion: Shame and pathological dissociation are important for not only the effect they have on the development of other complex PTSD symptoms, but also their direct and indirect effects on distress associated with relationships.

Dimethyl Acetals, an Indirect Marker of the Endogenous Antioxidant Plasmalogen Level, are Reduced in Blood Lipids of Sudanese Pre-eclamptic Subjects whose Background Diet is High in Carbohydrate

In Sudanese women with (n = 60) and without (n = 65) pre-eclampsia, circulating lipids, plasma and red cell saturated and monounsaturated fatty (MUFA) acids and dimethyl acetals (DMAs) were investigated. DMAs are an indirect marker of levels of plasmalogens, endogenous antioxidants, which play a critical role in oxidative protection, and cholesterol homeostasis. The pre-eclamptics had higher C18:1n-9 (p < 0.001) and ΣMUFA (p < 0.01) in plasma free fatty acids, C16:1n-7, C18:1n-9, ΣMUFA; 16:0/16:1n-7 (p < 0.01) in erythrocyte choline phosphoglycerides (ePC) and 16:1n-7, 18:1n-7 and 16:0/16:1n-7 (p < 0.01) in erythrocyte ethanolamine phosphoglycerides (ePE). In contrast, the DMAs 18:0, 18:1 and ΣDMAs in ePE, and 16:0, 18:0 and ΣDMAs in ePC were reduced (p < 0.001) in the pre-eclamptic women. This study of pregnant women with high carbohydrate and low fat background diet suggests pre-eclampsia is associated with oxidative stress and enhanced activity of the microsomal enzyme stearyl-CoA desaturase (delta 9 desaturase), as assessed by palmitic/palmitoleic (C16:0/C16:n-1) and stearic/oleic (C18/C18:1n-9) ratios.

Indirect Effects of Parental Coping on Children’s Psychosocial Adjustment in the Context of Pediatric Cancer

Thesis (Ph.D.)--University of Washington, 2016-08

Likelihood-Based Inference for Partially Observed Multi-Type Markov Branching Processes

Thesis (Ph.D.)--University of Washington, 2016-08