Since the Netflix Prize competition, latent factor models (LFMs) have become the comparison ``staples'' for many of the recent recommender methods. Meanwhile, it is still unclear to understand the impact of data preprocessing and updating algorithms on LFMs. The performance improvement of LFMs over baseline approaches, however, hovers at only low percentage numbers. Therefore, it is time for a better understanding of their real power beyond the overall root mean square error (RMSE), which as it happens, lies at a very compressed range, without providing too much chance for deeper insight. We introduce an experiment based handbook of LFMs and reveal data preprocessing and updating algorithms' power. We perform a detailed experimental study regarding the performance of classical staple LFMs on a classical dataset, Movielens 1M, that sheds light on a much more pronounced excellence of LFMs for particular categories of users and items, for RMSE and other measures. In particular, LFMs exhibit surprising and excellent advantages when handling several difficult user and item categories. By comparing the distributions of test ratings and predicted ratings, we show that the performance of LFMs is influenced by rating distribution. We then propose a method to estimate the performance of LFMs for a given rating dataset. Also, we provide a very simple, open-source library that implements staple LFMs achieving a similar performance as some very recent (2013) developments in LFMs, and at the same time being more transparent than some other libraries in wide use. / Graduate

This study performs a stress testing exercise on the Italian banking system in view of the 2007 financial crisis which was triggered by the crash of subprime mortgages. At the base of the global financial crisis was a failure of finan- cial regulators to quantify the accumulation of endogenous risks. Following the crisis, stress testing has acquired particular emphasis in the field of risk measurement under the Basel II supervisory framework. An econometric rela- tionship between the probability of default and the macroeconomic indicators is modeled according to the Merton approach for structural analysis using data on the Italian banking system. A latent factor model is employed to under- stand the dependence of the credit risk on the changes in the macroeconomic environment. The resulting relationship is exploited to compute the capital requirement under stressed conditions in order to draw inference about the resilience of the Italian banking system. JEL Classification G0, G01, G17, G10, C50, C22 Keywords Financial crisis, macroeconomic stress testing, credit risk, latent-factor model Author's e-mail jacomessi@yahoo.it Supervisor's e-mail petr.gapko@seznam.cz Abstrakt Klasifikace JEL G0, G01, G17, G10, C50, C22 Klíčová slova Financial crisis, macroeconomic stress test- ing, credit risk,...

abstract: Current models of pain coping typically focus on how pain contributes to poor physical and psychological functioning. Researchers have argued that this focus on the negative consequences is too narrow and does not account for times when individuals are able to maintain meaningful functioning despite their pain. Thus, the current study sought to investigate the day-to-day processes that both help and hinder recovery from pain and persistence towards daily goals. Specifically, the present study tested: a) a two-factor model of risk and resilience “factors” that capture key processes across affective, cognitive and social dimensions of functioning, and b) whether the relation between morning pain and end-of-day physical disability is mediated by increases in these afternoon risk and resilience factors. Within-day study measures were collected for 21 days via an automated phone system from 220 participants with Fibromyalgia. The results of multi-level confirmatory factor analysis indicated that, consistent with prediction, risk and resilience do constitute two factors. Findings from multilevel structural equation models also showed resilience factor mediated the link between late morning increases in pain and end-of-day disability, in line with hypotheses. Although the vulnerability factor as a whole did not mediate the within-day link between pain and disability, pain-catastrophizing individually did serve as a significant mediator of this relation. This study was the first to empirically test a within-day latent factor model of resilience and vulnerability and the first to capture the multidimensional nature of the pain experience by examining mechanisms across affective, cognitive and social domains of functioning. The findings of the current study suggest that in addition to studying the processes by which pain has a negative influence on the lives of pain sufferers, our understanding of the pain adaptation process can be further improved by concurrently examining mechanisms that motivate individuals to overcome the urge to avoid pain and to function meaningfully despite it. / Dissertation/Thesis / Doctoral Dissertation Psychology 2018

The objective of this study is to investigate consumers in the Taiwan region, the situation that when their due rights were being infringed, they had rather accept the unfair treatment from the manufacturers or suppliers, and will not strive for their own rights. In the consumer market, events of consumer right infringement are happening each day, seriously hindering the market order of fair competition. In this moment of the 2007, what are the thoughts within the minds of the consumers in Taiwan ? What are the factors that influence consumers striving for their due rights? In this study, the following issues were reviewed: relationships between manufacturers and consumers; consumer¡¦s cognizance of consumer rights; consumer protection; the roles of the law; government and consumer protection institutions in consumer protection; consumer education; and, consumer self-protection of consumer rights. A survey questionnaire was developed based on five themes of consumers themselves, consumer knowledge, law, government and consumer protection institutions. The survey attempts to understand the internal views of consumers. Consumers in the northern, central and southern Taiwan were randomly sampled according to population distribution. After collecting 170 questionnaires, the responses were coded and analyzed with SAS (statistical software) using Factor Analysis, one-way MANOVA and one-way ANOVA. Several latent factors were extracted, and the difference between consumer gender, age, education background and living region were studied. The results of statistical analysis indicate in 2007, the four main factors affecting consumers¡¦ strive for their rights are: (1) lack of external protection, (2) lack of self-confidence in claiming their rights, (3) dysfunction of consumer protection institutions, and, (4) lack of consumer knowledge. The results further show that the factors differ among living regions, however there is no evidence that there are differences in consumers gender, age and education background. This study has also investigates the level of consumer rights awareness, and the differences in gender, age, education background and living region in such cognizance. The results of statistical analysis show a very low awareness of consumer rights, and there is no evidence to conclude difference between gender, age, education background and living region.

<p>Current frontiers in complex stochastic modeling of high-dimensional processes include major emphases on so-called functional data: problems in which the data are snapshots of curves and surfaces representing fundamentally important scientific quantities. This thesis explores new Bayesian methodologies for functional data analysis. </p><p>The first part of the thesis places emphasis on the role of factor models in functional data analysis. Data reduction becomes mandatory when dealing with such high-dimensional data, more so when data are available on a large number of individuals. In Chapter 2 we present a novel Bayesian framework which employs a latent factor construction to represent each variable by a low dimensional summary. Further, we explore the important issue of modeling and analyzing the relationship of functional data with other covariate and outcome variables simultaneously measured on the same subjects.</p><p>The second part of the thesis is concerned with the analysis of circadian data. The focus is on the identification of circadian genes that is, genes whose expression levels appear to be rhythmic through time with a period of approximately 24 hours. While addressing this goal, most of the current literature does not account for the potential dependence across genes. In Chapter 4, we propose a Bayesian approach which employs latent factors to accommodate dependence and verify patterns and relationships between genes, while representing the true gene expression trajectories in the Fourier domain allows for inference on period, phase, and amplitude of the signal.</p><p>The third part of the thesis is concerned with the statistical analysis of computer models (simulators). The heavy computational demand of these input-output maps calls for statistical techniques that quickly estimate the surface output at untried inputs given a few preliminary runs of the simulator at a set design points. In this regard, we propose a Bayesian methodology based on a non-stationary Gaussian process. Relying on a model-based assessment of uncertainty, we envision a sequential design technique which helps choosing input points where the simulator should be run to minimize the uncertainty in posterior surface estimation in an optimal way. The proposed non-stationary approach adapts well to output surfaces of unconstrained shape.</p> / Dissertation

abstract: This dissertation comprises three chapters. In chapter one, using a rich dataset for the United States, I estimate a series of models to document the birth order effects on cognitive outcomes, non-cognitive outcomes, and parental investments. I estimate a model that allows for heterogeneous birth order effects by unobservables to examine how birth order effects varies across households. I find that first-born children score 0.2 of a standard deviation higher on cognitive and non-cognitive outcomes than their later-born siblings. They also receive 10\% more in parental time, which accounts for more than half of the differences in outcomes. I document that birth order effects vary between 0.1 and 0.4 of a standard deviation across households with the effects being smaller in households with certain characteristics such as a high income. In chapter two, I build a model of intra-household resource allocation that endogenously generates the decreasing birth order effects in household income with the aim of using the model for counterfactual policy experiments. The model has a life-cycle framework in which a household with two children confronts a sequence of time constraints and a lifetime monetary constraint, and divides the available time and monetary resources between consumption and investment. The counterfactual experiment shows that an annual income transfer of 10,000 USD to low-income households decreases the birth order effects on cognitive and non-cognitive skills by one-sixth, which is five times bigger than the effect in high-income household. In chapter three, with Francesco Agostinelli and Matthew Wiswall, we examine the relative importance of investments at home and at school during an important transition for many children, entering formal schooling at kindergarten. Moreover, our framework allows for complementarities between children's skills and investments from schools. We find that investments from schools are an important determinant of children's skills at the end of kindergarten, whereas parental investments, although strongly correlated with end-of-kindergarten outcomes, have smaller effects. In addition, we document a negative complementarity between children's skills at kindergarten entry and investments from schools, implying that low-skill children benefit the most from an increase in the quality of schools. / Dissertation/Thesis / Doctoral Dissertation Economics 2018

The significance of credit risk models has increased with the introduction of new Basel accord known as Basel II. The aim of this study is default rate modeling. This thesis follows the two possible approaches of a macro credit risk modeling. First, empirical models are investigated. Second, a latent factor model based on Merton's idea is introduced. Both of these models are derived from individual default probability models. We employed data over the time period from 1988 to 2003 of the Finnish economy in the first part of this thesis. Time series of bankruptcy and firm's numbers were used. Aggregate data for whole economy as well as industry specific data were available. First, linear vector autoregressive models was used in case of dynamic empirical model. We examined how significant macroeconomic indicators determined the default rate in the whole economy and in the industry specific sector. However these models cannot provide microeconomic foundation as latent factor models. We employed a one- factor model in our estimation although, multi-factor models were also considered. A one-factor model was estimated using disaggregated industrial data. This estimation can help understand relation between credit risk and macroeconomic indicators. Obtained results were used in the second part of this...

A latent factor model for count data is popularly applied when deconvoluting mixed signals in biological data as exemplified by sequencing data for transcriptome or microbiome studies. Due to the availability of pure samples such as single-cell transcriptome data, the estimators can enjoy much better accuracy by utilizing the extra information. However, such an advantage quickly disappears in the presence of excessive zeros. To correctly account for such a phenomenon, we propose a zero-inflated non-negative matrix factorization that models excessive zeros in both mixed and pure samples and derive an effective multiplicative parameter updating rule. In simulation studies, our method yields smaller bias comparing to other deconvolution methods. We applied our approach to gene expression from brain tissue as well as fecal microbiome datasets, illustrating the superior performance of the approach. Our method is implemented as a publicly available R-package, iNMF. In zero-inflated non-negative matrix factorization (iNMF) for the deconvolution of mixed signals of biological data, pure-samples play a significant role by solving the identifiability issue as well as improving the accuracy of estimates. One of the main issues of using single-cell data is that the identities(labels) of the cells are not given. Thus, it is crucial to sort these cells into their correct types computationally. We propose a nonlinear latent variable model that can be used for sorting pure-samples as well as grouping mixed-samples via deep neural networks. The computational difficulty will be handled by adopting a method known as variational autoencoding. While doing so, we keep the NMF structure in a decoder neural network, which makes the output of the network interpretable.

Intrinsic connectivity networks (ICNs) refer to brain functional networks that are consistently found under various conditions, during tasks or at rest. Some studies demonstrated that while some stimuli do not impact intrinsic connectivity, other stimuli actually activate intrinsic connectivity through suppression, excitation, moderation or modi cation. Most analyses of functional magnetic resonance imaging (fMRI) data use ad-hoc methods to estimate the latent structure of ICNs. Modeling the effects on ICNs has also not been fully investigated. Bayesian Intrinsic Connectivity Network (BICNet) captures the ICN structure with We propose a BICNet model, an extended Bayesian dynamic sparse latent factor model, to identify the ICNs and quantify task-related effects on the ICNs. BICNet has the following advantages: (1) It simultaneously identifies the individual and group-level ICNs; (2) It robustly identifies ICNs by jointly modeling resting-state fMRI (rfMRI) and task-related fMRI (tfMRI); (3) Compared to independent component analysis (ICA)-based methods, it can quantify the difference of ICNs amplitudes across different states; (4) The sparsity of ICNs automatically performs feature selection, instead of ad-hoc thresholding. We apply BICNet to the rfMRI and language tfMRI data from the Human Connectome Project (HCP) and identify several ICNs related to distinct language processing functions.

As pesquisas científicas na área da Psiquiatria freqüentemente avaliam características subjetivas de indivíduos como, por exemplo, depressão, ansiedade e fobias. Os dados são coletados através de questionários, cujos itens tentam identificar a presença ou ausência de certos sintomas associados à morbidade psiquiátrica de interesse. Alguns desses itens, entretanto, podem provocar constrangimento em parte dos indivíduos respondedores por abordarem características ou comportamentos socialmente questionáveis ou, até, ilegais. Um modelo da teoria de resposta ao item é proposto neste trabalho visando diferenciar a relação entre a probabilidade de presença do sintoma e a gravidade da morbidade de indivíduos constrangidos e não constrangidos. Itens que necessitam dessa diferenciação são chamados \\textbf{itens com comportamento diferencial}. Adicionalmente, o modelo permite assumir que indivíduos constrangidos em responder um item possam vir a mentir em suas respostas, no sentido de omitir a presença de um sintoma. Aplicações do modelo proposto a dados simulados para questionários com 20 itens mostraram que as estimativas dos parâmetros são próximas aos seus verdadeiros valores. A qualidade das estimativas piora com a diminuição da amostra de indivíduos, com o aumento do número de itens com comportamento diferencial e, principalmente, com o aumento do número de itens com comportamento diferencial suscetíveis à mentira. A aplicação do modelo a um conjunto de dados reais, coletados para avaliar depressão em adolescentes, ilustra a diferença do padrão de resposta do item ``crises de choro\" entre homens e mulheres. / Psychiatric scientific research often evaluate subjective characteristics of the individual such as depression, anxiety and phobias. Data are collected through questionnaires with items that try to identify the presence or absence of certain symptoms associated with the psychiatric disease. Some of these items though could make some people embarrassed since they are related to questionable or even illegal social behaviors. The item response theory model proposed within this work envisions to differentiate the relationship between the probability of the symptom presence and the gravity of the disease of embarrassed and non-embarrassed individuals. Items that need this differentiation are called differential item functioning (dif). Additionally, the model has the assumption that individuals embarrassed with one particular item could lie across other answers to omit a possible condition. Applications of the proposed model to simulated data for a 20-item questionnaire have showed that parameter estimates of the proposed model are close to their real values. The estimate accuracy gets worse as the number of individuals decreases, the number of dif increases, and especially as the number of dif susceptible to lying increases. The application of the model to a group of real data, collected to evaluate teenager depression, shows the difference in the probability of \"crying crisis\" presence between men and women.