Wellness and healthy life are the most common concerns for an individual to lead a happy life. A web-based approach known as Wellness Scoring is being developed taking into people’s concerns for their health issues. In this approach, four different classifiers are being investigated to predict the wellness. In this thesis, we investigated four different classifiers (a probabilistic graphical model, simple probabilistic classifier, probabilistic statistical classification and an artificial neural network) to predict the wellness outcome. An approach to calculate wellness score is also addressed. All these classifiers are trained on real data, hence giving more accurate results. With this solution, there is a better way of keeping track of an individuals’ health. In this thesis, we present the design and development of such a system and evaluate the performance of the classifiers and design considerations to maximize the end user experience with the application. A user experience model capable of predicting the wellness score for a given set of risk factors is developed. / Includes bibliography. / Thesis (M.S.)--Florida Atlantic University, 2014. / FAU Electronic Theses and Dissertations Collection

Neste trabalho discutimos os principais argumentos da inferência bayesiana subjetivista. Posteriormente, a partir de uma revisão da literatura dos testes de hipóteses, os principais testes são analisados sob a ótica da teoria da decisão, particularmente no que tange às hipóteses precisas. Adicionalmente, funções de perda para testes de significância, seguindo a proposta de Fisher e do FBST, são analisadas e comparadas. / This work discusses the main points of the bayesian subjectivist inference. Posteriorly, from a literature review of hypothesis testing, the main approaches are interpreting from a decision-theoretic viewpoint, particularly regarding the precise hypotheses. Additionally, loss functions for tests of significance, following the proposal of Fisher and FBST, are analyzed and compared.

Mapping of mortality rates has been a valuable public health tool. We describe novel Bayesian methods for constructing maps which do not depend on a post stratification of the estimated rates. We also construct posterior modal maps rather than posterior mean maps. Our methods are illustrated using mortality data from chronic obstructive pulmonary diseases (COPD) in the continental United States. Poisson regression models have attracted much attention in the scientific community for their superiority in modeling rare events (including mortality counts from COPD). Christiansen and Morris (JASA 1997) described a hierarchical Bayesian model for heterogeneous Poisson counts under the exchangeability assumption. We extend this model to include latent classes (groups of similar Poisson rates unknown to an investigator). Also, it is standard practice to construct maps using quantiles (e.g., quintiles) of the estimated mortality rates. For example, based on quintiles, the mortality rates are cut into 5 equal size groups, each containing $20\%$ of the data, and a different color is applied to each of them on the map. A potential problem is that, this method assumes an equal number of data in each group, but this is often not the case. The latent class model produces a method to construct maps without using quantiles, providing a more natural representation of the colors. Typically, for rare events, the posterior densities of the rates are skewed, making the posterior mean map inappropriate and inaccurate. Thus, although it is standard practice to present the posterior mean maps, we also develop a method to provide the joint posterior modal map (i.e., the map with the highest posterior probability over the ensemble). For the COPD data, collected 1988-1992 over 798 health service areas, we use Markov chain Monte Carlo methods to fit the model, and an output analysis is used to construct the new maps.

We have considered the problem in which a biased sample is selected from a finite population, and this finite population itself is a random sample from an infinitely large population, called the superpopulation. The parameters of the superpopulation and the finite population are of interest. There is some information about the selection mechanism in that the selection probabilities are linearly related to the measurements. This is typical of establishment surveys where the selection probabilities are taken to be proportional to the previous year's characteristics. When all the selection probabilities are known, as in our problem, inference about the finite population can be made, but inference about the distribution is not so clear. For continuous measurements, one might assume that the the values are normally distributed, but as a practical issue normality can be tenuous. In such a situation a transformation to normality may be useful, but this transformation will destroy the linearity between the selection probabilities and the values. The purpose of this work is to address this issue. In this light we have constructed two models, an ignorable selection model and a nonignorable selection model. We use the Gibbs sampler and the sample importance re-sampling algorithm to fit the nonignorable selection model. We have emphasized estimation of the finite population parameters, although within this framework other quantities can be estimated easily. We have found that our nonignorable selection model can correct the bias due to unequal selection probabilities, and it provides improved precision over the estimates from the ignorable selection model. In addition, we have described the case in which all the selection probabilities are unknown. This is useful because many agencies (e.g., government) tend to hide these selection probabilities when public-used data are constructed. Also, we have given an extensive theoretical discussion on Poisson sampling, an underlying sampling scheme in our models especially useful in the case in which the selection probabilities are unknown.

It is customary when presenting a choropleth map of rates or counts to present only the estimates (mean or mode) of the parameters of interest. While this technique illustrates spatial variation, it ignores the variation inherent in the estimates. We describe an approach to present variability in choropleth maps by constructing 100(1-alpha)% simultaneous intervals. The result provides three maps (estimate with two bands). We propose two methods to construct simultaneous intervals from the optimal individual highest posterior density (HPD) intervals to ensure joint simultaneous coverage of 100(1-alpha)%. Both methods exhibit the main feature of multiplying the lower bound and dividing the upper bound of the individual HPD intervals by parameters 0

Orientador: José Augusto Chinellato / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Física Gleb Wataghin / Made available in DSpace on 2018-09-03T07:30:39Z (GMT). No. of bitstreams: 1 Perassa_EderArnedo_D.pdf: 3556275 bytes, checksum: c4e6180df4a4a5dcbfe476b7d331bee4 (MD5) Previous issue date: 2017 / Resumo: Neste trabalho, estudamos os métodos de estatística bayesiana e máxima entropia na análise de dados em eventos de raios cósmicos. Inicialmente, fizemos um resumo sobre o desenvolvimento da física de raios cósmicos em que descrevemos alguns resultados teóricos e experimentais recentes. A seguir, apresentamos uma breve revisão do método bayesiano e o aplicamos na determinação da composição em massa dos primários em eventos de raios cósmicos. Além disso, introduzimos o método de máxima entropia e propomos um método de parametrização do perfil longitudinal de chuveiros atmosféricos extensos. Em todas as aplicações, foram mostrados os algoritmos desenvolvidos e os resultados obtidos a partir de dados de eventos simulados. Os resultados indicaram que tais métodos podem ser utilizados satisfatoriamente como ferramentas na análise de dados em eventos de raios cósmicos / Abstract: In this work, we study bayesian statistics and maximum entropy methods in cosmic ray events data analysis. At first, we summarized developments in cosmic rays physics, describing some recent theoretical and experimental results. We present briefly a review of bayesian method and apply it to the problem of determining mass composition primary cosmic ray events. Moreover, we introduce the maximum entropy method and propose a method for the parametrization of the longitudinal profile of extensive air showers. In all applications, the algorithms developed and the results obtained from simulated event data were shown. The results suggested that such methods can be satisfactorily used as tools in cosmic rays events data analysis / Doutorado / Física / Doutor em Ciências / 277612/2007 / CAPES

Durant les dernières décennies, deux théories a priori contradictoires l’une avec l’autre,prétendent donner un fondement aux comportements des agents économiques. La théorie de la décision axiomatique, la plus ancienne cherche à décrire les comportements à partir du principe de rationalité, alors que l’économie comportementale se base principalement sur une analyse empirique et expérimentale. Cette thèse, prend le parti de réunir ces deux points de vues en mobilisant le concept de préférence incomplète. Leurs fléxibilités capturent de nombreux comportements observés lors des expériences, et leurs structures riches permettent une analyse normative. Dans cette optique, nous développons dans la première partie un modèle d’aversion au risque dynamique, en modélisant la notion de bienêtre par une relation de préférence incomplète. Nous montrons que le bien-être du décideur est représenté par deux psychologies contradictoires. La première traduit l’aversion au risque sur le long terme et, est représentée par le modèle de l’espérance d’utilité, la deuxième décrit une réaction plus émotionnelle face au risque, et est caractérisée par le modèle de l’espérance d’utilité au sens de Choquet. Dans la seconde partie, nous démontrons quelles sont les conditions comportementales, nécessaires et suffisantes permettant à une relation de préférences incomplète d’être représentée par l’intersection d’un ensemble de relation de préférences complètes vérifiant l’axiome de l’indépendance comonotone. / In recent decades, two theories which seems contradictory, claim that they can provide abasis for the behavior of economic agents, i.e the theory of decision and the behavioral economics. We have tried, in this thesis to unite these two points of view by mobilizing the concept of incomplete preference. We develop in the first part a model of time varying risk aversion: we show that the Decision Maker anticipates that the passage of time will have an effect on him outlook. By modeling the notion of well-being with a incomplete preference,we show that the welfare of the decision maker is represented by two contradictory psychologies. The first reflects the risk aversion in the long term and is represented by the model of expected utility, the second describes a more emotional response to risk, and is characterized by the model of Choquet expected utility. In the second part, we identify the behavioral conditions, both necessary and sufficient, in which an incomplete preference relation could be represented by the intersection of a set of complete and transitive preference relation satisfying the axiom of comonotone independence.

Neste trabalho discutimos os principais argumentos da inferência bayesiana subjetivista. Posteriormente, a partir de uma revisão da literatura dos testes de hipóteses, os principais testes são analisados sob a ótica da teoria da decisão, particularmente no que tange às hipóteses precisas. Adicionalmente, funções de perda para testes de significância, seguindo a proposta de Fisher e do FBST, são analisadas e comparadas. / This work discusses the main points of the bayesian subjectivist inference. Posteriorly, from a literature review of hypothesis testing, the main approaches are interpreting from a decision-theoretic viewpoint, particularly regarding the precise hypotheses. Additionally, loss functions for tests of significance, following the proposal of Fisher and FBST, are analyzed and compared.

In this dissertation, I propose that, post-action, people tend to be more optimistic about outcomes when their actions were based on malleable (vague) information compared to when their actions were based on unmalleable (precise) information. However, pre-action, no such difference occurs. I term this inconsistency in optimism in the pre and post-action stage, the Information Malleability Effect (IME). These actions could include the choice of a product, drawing a ball from an urn, or consumption of a food item. Prior research on ambiguity aversion has reliably documented that people are generally averse to making decisions based on malleable information. On the other hand, research on situated optimism has demonstrated that people exhibit a high level of optimism for events they consider more desirable and they distort the available information to make the desirable events seem more likely to occur. I review these two streams of literature and show that although both literatures make predictions in either the pre or the post-action stage, neither of them alone can explain the IME. I propose a theoretical framework to explain the underlying cause of the IME that combines these two streams of literature and utilizes the motivated reasoning account. Based on this framework, I posit hypotheses that are tested across a series of experiments. Experiment 1a and 1b demonstrate the IME in a between and within participant design. Experiment 2 demonstrates that interpretational flexibility of malleable information results in positive outcomes appearing more plausible and negative outcomes less plausible compared to when information is unmalleable. Experiment 3 provides support for the proposed underlying process by priming accuracy and desired goals.

En sciences économiques, l'une des questions centrales concerne l'allocation des ressources rares et plus particulièrement leur répartition. La décision apparait ainsi au cœur des thématiques économiques, que ce soit en micro-économie ou en macro-économie. Dans un premier temps, nous revenons sur le fait que l’information, élément central de la prise de décision, est imparfaite. En utilisant la théorie des ensembles flous, qui a pour objet de capturer l’imprécision, nous construisons un nombre flou nommé nombre flou C-Shape qui permet de capter la sensibilité du preneur de décision. Nous étudions ensuite la théorie de la décision au travers de deux axes de recherche à savoir (1) la recherche opérationnelle couplée à la théorie des jeux et (2) l’aide à la décision. En premier lieu, Nous faisons une analogie entre la fonction distance et la fonction d’appartenance. Grâce à l’hypothèse de B-convexité et à la fonction C-Shape nous construisons des classes de jeux pour lesquels les joueurs peuvent être optimistes, pessimistes ou neutres, et pour lesquels l’existence d’équilibre de Nash est avérée. Enfin, concernant l’aide à la décision, nous utilisons la fonction C-Shape pour caractériser un nouveau type de critère nommé pseudo critère C-Shape qui permet de considérer les alternatives comme substituables. Ceci permet de prendre en compte, par exemple le contexte institutionnel dans lequel la prise de décision est prise. / Determining the allocation and the distribution of scarce resources is fundamental in economics. Thus, decision theory is the cornerstone of economic theory. In this thesis, we first provide a state of the art insisting on the fact that information, that is a central element of decision-making, is imperfect. Secondly, using fuzzy set theory, which aims to capture imprecision, we construct a fuzzy number, so-called C-Shape that captures the sensitivity of the decision-maker. Thirdly, we study decision theory through two key concepts of operation research: (1) game theory and (2) multi-criteria decision making. We provide an analogy between the gauge functions of convex sets and the membership functions arising in fuzzy set theory. Coupling a suitable notion of -convexity with the C-Shape function, we introduce a class of games for which the players can be optimistic, pessimistic or neutral. In addition the existence of Nash equilibrium is proved for such a class of games. Finally, concerning multi-criteria decision analysis, we use the C-Shape functions to characterize a new type of criteria called C-Shape pseudo-criterion, which makes possible to consider the alternatives as substitutable. This should be of interest to take into account, for example, the institutional context in which decision-making is taken.