Doctor of Philosophy / Department of Psychological Sciences / Gary L. Brase / Operational Sex Ratio (OSR: the ratio of reproductively viable males to females in a given population) has been theorized and studied as a construct that may influence behaviors. The encoding of sex ratio was examined in order to determine whether the cognitive process underlying it is automatic or effortful. Further, the current work examines whether OSR or Adult Sex Ratio (ASR: the ratio of adult males to females) is encoded. The current work involved four experiments; two using frequency tracking methodology and two using summary statistic methodology. Experiment 1 found a strong correlation between OSR of conditions and estimates of sex ratio. Participants in Experiment 1 were uninformed on the purpose of the experiment, thus the strong correlations between actual and estimated sex ratio suggest a level of automaticity. Experiment 2 found a strong correlation between the ASR of conditions and estimates, suggesting that individuals do not encode OSR over ASR. Experiments 3.a. and 3.b. demonstrated automaticity in estimates of sex ratio from briefly presented sets of faces, for two different durations: 1000ms and 330ms, the later of which is widely accepted as the length of a single eye fixation. Overall this work demonstrated a human ability to recall proportion of sexes from arrays of serially presented individuals (Experiments 1 and 2), and that ASR is encoded when participants are presented with conditions including older adults. This work found the encoding of sex ratio to be highly automatic, particularly stemming from the results of Experiments 3.a. and 3.b. Conclusions from this work help to verify previous research on sex ratio’s effect on mating strategies through evidence supporting the automatic nature of encoding sex ratio. Further, the current work is a foundation for future research regarding sex ratio, and leads to several proposals for future endeavors.

Panel data, also known as longitudinal data, are composed of repeated measurements taken from the same subject over different time points. Although it is generally used in time series applications, forecasting can also be used in panel data due to its time dimension. However, there is limited number of studies in this area in the literature. In this thesis, forecasting is studied for panel data with binary response because of its increasing importance and increasing fundamental roles. A simulation study is held to compare the efficiency of different methods and to find the one that gives the optimal forecast values. In this simulation, 21 different methods, including na&iuml / ve and complex ones, are used by the help of R software. It is concluded that transition models and random effects models with no lag of response can be chosen for getting the most accurate forecasts, especially for the first two years of forecasting.

<div><div><div><p>Impacting individual’s social and physical well-being, psychiatric disorders have been a substantial burden on public health. As such disorders are frequently observed aggregating in families, we can expect a large involvement of heritable components underlying their etiologies. Therefore, studying the genetic architecture and basis is one of the most important aims toward developing effective treatments for psychiatric disorders. The overall objective of this dissertation is to contribute to understanding the genetics of psychiatric disorders. Analyzing summary statistics from genomewide association studies (GWAS) of psychiatric disorders, we mainly present results of two projects. In the first one, we evaluated commonalities and distinctions in genetic risk of four highly comorbid childhood onset neuropsychiatric disorders: attention deficit/hyperactivity disorder (ADHD), autism spectrum disorder (ASD), obsessive-compulsive disorder (OCD) and Tourette’s syndrome (TS). Through systematic analysis of genetic architecture and correlation, we confirmed exitance of genetic components shared across ADHD, ASD and TS, as well as OCD and TS. Subsequently, we identified those components at variant, gene, and tissue specificity levels through meta-analyses. Our results pointed toward possible involvement of hypothalamus-pituitary-adrenal (HPA) axis, a human stress response system, in the etiology of these childhood onset disorders. The second project includes the proposition of a novel framework for general GWAS summary statistics-based analyses. Instead of regular odds ratio and standard errors archived in the summary statistics, we proposed a recounstruction approach to rewrite the results in terms of single nucleotide polymorphisms (SNP) allelic and genotypic frequencies. We also put forward three applications built-upon the proposed framework, and evaluated the performance on both synthetic data and real GWAS results of psychiatric disorders for each of them. Through these three applications, we demonstrated that this framework can broaden the scope of GWAS summary statistics-based analyses and unify various of analyses pipelines. We hope our work can serve as a stepping-stone for future researchers aiming at understanding and utilizing GWAS results of complex psychiatric disorders.</p></div></div></div>

No description available.

Processos estocásticos complexos são muitas vezes utilizados em modelagem, com o intuito de capturar uma maior proporção das principais características dos sistemas biológicos. A descrição do comportamento desses sistemas tem sido realizada por muitos amostradores baseados na distribuição a posteriori de Monte Carlo. Modelos probabilísticos que descrevem esses processos podem levar a funções de verossimilhança computacionalmente intratáveis, impossibilitando a utilização de métodos de inferência estatística clássicos e os baseados em amostragem por meio de MCMC. A Computação Bayesiana Aproximada (ABC) é considerada um novo método de inferência com base em estatísticas de resumo, ou seja, valores calculados a partir do conjunto de dados (média, moda, variância, etc.). Essa metodologia combina muitas das vantagens da eficiência computacional de processos baseados em estatísticas de resumo com inferência estatística bayesiana uma vez que, funciona bem para pequenas amostras e possibilita incorporar informações passadas em um parâmetro e formar uma priori para análise futura. Nesse trabalho foi realizada uma comparação entre os métodos de estimação, clássico, bayesiano e ABC, para estudos de simulação de modelos simples e para análise de dados de dinâmica populacional. Foram implementadas no software R as distâncias modular e do máximo como alternativas de função distância a serem utilizadas no ABC, além do algoritmo ABC de rejeição para equações diferenciais estocásticas. Foi proposto sua utilização para a resolução de problemas envolvendo modelos de interação populacional. Os estudos de simulação mostraram melhores resultados quando utilizadas as distâncias euclidianas e do máximo juntamente com distribuições a priori informativas. Para os sistemas dinâmicos, a estimação por meio do ABC apresentou resultados mais próximos dos verdadeiros bem como menores discrepâncias, podendo assim ser utilizado como um método alternativo de estimação. / Complex stochastic processes are often used in modeling in order to capture a greater proportion of the main features of natural systems. The description of the behavior of these systems has been made by many Monte Carlo based samplers of the posterior distribution. Probabilistic models describing these processes can lead to computationally intractable likelihood functions, precluding the use of classical statistical inference methods and those based on sampling by MCMC. The Approxi- mate Bayesian Computation (ABC) is considered a new method for inference based on summary statistics, that is, calculated values from the data set (mean, mode, variance, etc.). This methodology combines many of the advantages of computatio- nal efficiency of processes based on summary statistics with the Bayesian statistical inference since, it works well for small samples and it makes possible to incorporate past information in a parameter and form a prior distribution for future analysis. In this work a comparison between, classical, Bayesian and ABC, estimation methods was made for simulation studies considering simple models and for data analysis of population dynamics. It was implemented in the R software the modular and maxi- mum as alternative distances function to be used in the ABC, besides the rejection ABC algorithm for stochastic differential equations. It was proposed to use it to solve problems involving models of population interaction. The simulation studies showed better results when using the Euclidean and maximum distances together with informative prior distributions. For the dynamic systems, the ABC estimation presented results closer to the real ones as well as smaller discrepancies and could thus be used as an alternative estimation method.

Processos estocásticos complexos são muitas vezes utilizados em modelagem, com o intuito de capturar uma maior proporção das principais características dos sistemas biológicos. A descrição do comportamento desses sistemas tem sido realizada por muitos amostradores baseados na distribuição a posteriori de Monte Carlo. Modelos probabilísticos que descrevem esses processos podem levar a funções de verossimilhança computacionalmente intratáveis, impossibilitando a utilização de métodos de inferência estatística clássicos e os baseados em amostragem por meio de MCMC. A Computação Bayesiana Aproximada (ABC) é considerada um novo método de inferência com base em estatísticas de resumo, ou seja, valores calculados a partir do conjunto de dados (média, moda, variância, etc.). Essa metodologia combina muitas das vantagens da eficiência computacional de processos baseados em estatísticas de resumo com inferência estatística bayesiana uma vez que, funciona bem para pequenas amostras e possibilita incorporar informações passadas em um parâmetro e formar uma priori para análise futura. Nesse trabalho foi realizada uma comparação entre os métodos de estimação, clássico, bayesiano e ABC, para estudos de simulação de modelos simples e para análise de dados de dinâmica populacional. Foram implementadas no software R as distâncias modular e do máximo como alternativas de função distância a serem utilizadas no ABC, além do algoritmo ABC de rejeição para equações diferenciais estocásticas. Foi proposto sua utilização para a resolução de problemas envolvendo modelos de interação populacional. Os estudos de simulação mostraram melhores resultados quando utilizadas as distâncias euclidianas e do máximo juntamente com distribuições a priori informativas. Para os sistemas dinâmicos, a estimação por meio do ABC apresentou resultados mais próximos dos verdadeiros bem como menores discrepâncias, podendo assim ser utilizado como um método alternativo de estimação. / Complex stochastic processes are often used in modeling in order to capture a greater proportion of the main features of natural systems. The description of the behavior of these systems has been made by many Monte Carlo based samplers of the posterior distribution. Probabilistic models describing these processes can lead to computationally intractable likelihood functions, precluding the use of classical statistical inference methods and those based on sampling by MCMC. The Approxi- mate Bayesian Computation (ABC) is considered a new method for inference based on summary statistics, that is, calculated values from the data set (mean, mode, variance, etc.). This methodology combines many of the advantages of computatio- nal efficiency of processes based on summary statistics with the Bayesian statistical inference since, it works well for small samples and it makes possible to incorporate past information in a parameter and form a prior distribution for future analysis. In this work a comparison between, classical, Bayesian and ABC, estimation methods was made for simulation studies considering simple models and for data analysis of population dynamics. It was implemented in the R software the modular and maxi- mum as alternative distances function to be used in the ABC, besides the rejection ABC algorithm for stochastic differential equations. It was proposed to use it to solve problems involving models of population interaction. The simulation studies showed better results when using the Euclidean and maximum distances together with informative prior distributions. For the dynamic systems, the ABC estimation presented results closer to the real ones as well as smaller discrepancies and could thus be used as an alternative estimation method.

No description available.

No description available.

Multi-armed bandit (MAB) algorithms could be used to select a subset of the k most informative summary statistics, from a pool of m possible summary statistics, by reformulating the subset selection problem as a MAB problem. This is suggested by experiments that tested five MAB algorithms (Direct, Halving, SAR, OCBA-m, and Racing) on the reformulated problem and comparing the results to two established subset selection algorithms (Minimizing Entropy and Approximate Sufficiency). The MAB algorithms yielded errors at par with the established methods, but in only a fraction of the time. Establishing MAB algorithms as a new standard for summary statistics subset selection could therefore save numerous scientists substantial amounts of time when selecting summary statistics for approximate bayesian computation.

With the increasing importance of hand-held devices with touch displays, the need for flat panel displays (FPDs) will likely increase in the future. Glass is the most important substrate for FPD manufacturing, where both its bulk and surface properties are critical for its performance. Many properties of the glass used in FPDs are controlled by its surface chemistry. Surface hydroxyls are the most important functional groups on a glass surface, which control processes that occurs on oxide surfaces, including wetting, adhesion, electrostatic charging and discharge, and the rate of contamination. In this dissertation, I present a new approach for determining surface silanol densities on planar surfaces. This methodology consists of tagging surface silanols using atomic layer deposition (ALD) followed by low energy ion scattering (LEIS) analysis of the tags. The LEIS signal is limited to the outermost atomic layer, i.e., LEIS is an extremely surface sensitive technique. Quantification in LEIS is straightforward in the presence of suitable reference materials. An essential part of any LEIS measurement is the preparation and characterization of the sample and appropriate reference materials that best represent the samples. My tag-and-count method was applied to chemically and thermally treated fused silica. In this work, I determined the silanol density of a fully hydroxylated fused silica surface to be 4.67 OH/nm2. This value agrees with the literature value for high surface area silica powder. My methodology should be important in future glass studies. Surface Science Spectra (SSS) is an important, peer-reviewed database of spectra from surfaces. Recently, SSS has been expanding to accept spectra from new surface techniques. I created the first SSS submission form for LEIS spectra (see appendix 5), and used it to create the first SSS LEIS paper (on CaF2 and Au reference materials, see chapter 3). I also show LEIS reference spectra for ZnO, and copper in the appendix 1. The rest of my dissertation focuses on my chemometrics/informatics and data analysis work. For example, I showed the performance and capabilities of a series of summary statistics as new tools for unsupervised exploratory data analysis (EDA) (see chapter 4). Unsupervised EDA is often the first step in understanding complex data sets because it can group, and even classify, samples according to their spectral similarities and differences. Pattern recognition entropy (PRE) and other summary statistics are direct methods for analyzing data - they are not factor-based approaches like principal component analysis (PCA) or multivariate curve resolution (MCR). I show that, in general, PRE outperforms the other summary statistics, especially in image analysis, although I recommend a suite of summary statistics be used in exploring complex data sets. In addition, I introduce the concept of divided spectrum-PRE (DS-PRE) as a new EDA method and use it to analyze multiple data sets. DS-PRE increases the discrimination power of PRE. I have also prepared a guide that discusses the vital aspects and considerations for chemometrics/informatics analyses of XPS data along with specific EDA tools that can be used to probe XPS data sets, including PRE, PCA, MCR, and cluster analysis (see chapter 5). I emphasize the importance of an initial evaluation/plotting of raw data, data preprocessing, returning to the original data after a chemometrics/informatics analysis, and determining the number of abstract factors to keep in an analysis, including reconstructing the data using PCA. In my thesis, I also show the analysis of commercial automotive lubricant oils (ALOs) with various chemometrics techniques (see chapter 6). Using these methods, the ALO samples were readily differentiated according to their American Petroleum Institute (API) classification and base oil types: mineral, semi-synthetic, and synthetic.