Prediciton of emergency events: a multi-task multi-label learning approach

Prediction of patient outcomes is critical to plan resources in an hospital emergency department. We present a method to exploit longitudinal data from Electronic Medical Records (EMR), whilst exploiting multiple patient outcomes. We divide the EMR data into segments where each segment is a task, and all tasks are associated with multiple patient outcomes over a 3, 6 and 12 month period. We propose a model that learns a prediction function for each task-label pair, interacting through two subspaces: the first subspace is used to impose sharing across all tasks for a given label. The second subspace captures the task-specific variations and is shared across all the labels for a given task. The proposed model is formulated as an iterative optimization problems and solved using a scalable and efficient Block co-ordinate descent (BCD) method. We apply the proposed model on two hospital cohorts - Cancer and Acute Myocardial Infarction (AMI) patients collected over a two year period from a large hospital emergency department. We show that the predictive performance of our proposed models is significantly better than those of several state-of-the-art multi-task and multi-label learning methods.

Securing sensor networks by moderating frequencies

Security of Wireless Sensor Network (WSN) is a key issue in information security. Most existing security protocols exploit various Mathematical tools to strengthen their security. Some protocols use the details of the geographical location of the nodes. However, to the best authors’ knowledge, none of the existing works exploit the constraints faced by the adversary, specifically, tracing a particular frequency from a large range of unknown frequency channels. The current work uses positional details of the individual nodes. Then the aim is to exploit this weakness of tracing frequencies by assigning a wide range of frequency channels to each node. Experiments using Magneto Optic Sensors reveal that any change of the parametric Faraday’s rotational angle affects the frequency of the Optical waves. This idea can perhaps be generalized for practically deployable sensors (having respective parameters) along with a suitable key management scheme.

Locally sparsified compressive sensing in magnetic resonance imaging

Magnetic Resonance Imaging (MRI) is a widely used technique for acquiring images of human organs/tissues. Due to its complex imaging process, it consumes a lot of time to produce a high quality image. Compressive Sensing (CS) has been used by researchers for rapid MRI. It uses a global sparsity constraint with variable density random sampling and L1 minimisation. This work intends to speed up the imaging process by exploiting the non-uniform sparsity in the MR images. Locally Sparsified CS suggests that the image can be even better sparsified by applying local sparsity constraints. The image produced by local CS can further reduce the sample set. This paper establishes the basis for a methodology to exploit non-uniform nature of sparsity and to make the MRI process time efficient by using local sparsity constraints.

A matrix factorization framework for jointly analyzing multiple nonnegative data sources

Nonnegative matrix factorization based methods provide one of the simplest and most effective approaches to text mining. However, their applicability is mainly limited to analyzing a single data source. In this chapter, we propose a novel joint matrix factorization framework which can jointly analyze multiple data sources by exploiting their shared and individual structures. The proposed framework is flexible to handle any arbitrary sharing configurations encountered in real world data. We derive an efficient algorithm for learning the factorization and show that its convergence is theoretically guaranteed. We demonstrate the utility and effectiveness of the proposed framework in two real-world applications—improving social media retrieval using auxiliary sources and cross-social media retrieval. Representing each social media source using their textual tags, for both applications, we show that retrieval performance exceeds the existing state-of-the-art techniques. The proposed solution provides a generic framework and can be applicable to a wider context in data mining wherever one needs to exploit mutual and individual knowledge present across multiple data sources.

Hierarchical Bayesian nonparametric models for knowledge discovery from electronic medical records

Electronic Medical Record (EMR) has established itself as a valuable resource for large scale analysis of health data. A hospital EMR dataset typically consists of medical records of hospitalized patients. A medical record contains diagnostic information (diagnosis codes), procedures performed (procedure codes) and admission details. Traditional topic models, such as latent Dirichlet allocation (LDA) and hierarchical Dirichlet process (HDP), can be employed to discover disease topics from EMR data by treating patients as documents and diagnosis codes as words. This topic modeling helps to understand the constitution of patient diseases and offers a tool for better planning of treatment. In this paper, we propose a novel and flexible hierarchical Bayesian nonparametric model, the word distance dependent Chinese restaurant franchise (wddCRF), which incorporates word-to-word distances to discover semantically-coherent disease topics. We are motivated by the fact that diagnosis codes are connected in the form of ICD-10 tree structure which presents semantic relationships between codes. We exploit a decay function to incorporate distances between words at the bottom level of wddCRF. Efficient inference is derived for the wddCRF by using MCMC technique. Furthermore, since procedure codes are often correlated with diagnosis codes, we develop the correspondence wddCRF (Corr-wddCRF) to explore conditional relationships of procedure codes for a given disease pattern. Efficient collapsed Gibbs sampling is derived for the Corr-wddCRF. We evaluate the proposed models on two real-world medical datasets - PolyVascular disease and Acute Myocardial Infarction disease. We demonstrate that the Corr-wddCRF model discovers more coherent topics than the Corr-HDP. We also use disease topic proportions as new features and show that using features from the Corr-wddCRF outperforms the baselines on 14-days readmission prediction. Beside these, the prediction for procedure codes based on the Corr-wddCRF also shows considerable accuracy.

A Diagnostic Assessment Of Evolutionary Multiobjective Optimization For Water Resources Systems

This study contributes a rigorous diagnostic assessment of state-of-the-art multiobjective evolutionary algorithms (MOEAs) and highlights key advances that the water resources field can exploit to better discover the critical tradeoffs constraining our systems. This study provides the most comprehensive diagnostic assessment of MOEAs for water resources to date, exploiting more than 100,000 MOEA runs and trillions of design evaluations. The diagnostic assessment measures the effectiveness, efficiency, reliability, and controllability of ten benchmark MOEAs for a representative suite of water resources applications addressing rainfall-runoff calibration, long-term groundwater monitoring (LTM), and risk-based water supply portfolio planning. The suite of problems encompasses a range of challenging problem properties including (1) many-objective formulations with 4 or more objectives, (2) multi-modality (or false optima), (3) nonlinearity, (4) discreteness, (5) severe constraints, (6) stochastic objectives, and (7) non-separability (also called epistasis). The applications are representative of the dominant problem classes that have shaped the history of MOEAs in water resources and that will be dominant foci in the future. Recommendations are provided for which modern MOEAs should serve as tools and benchmarks in the future water resources literature.

Liquidez e fluxo de caixa: um estudo teórico sobre alguns elementos que atuam no processo de formação do caixa e na determinação do nível de liquidez de empresas privadas não financeiras

O trabalho procura mapear e interpretar o processo de formação de caixa decompondo e analisando as atividades que contribuem para liberar ou retirar recursos do fluxo de caixa. Procura também avançar no problema da determinação do nível ótimo de liquidez que deve ser mantido pelas empresas.

Correio híbrido: um ensaio prospectivo sobre o posicionamento estratégico da ECT

A Empresa Brasileira de Correios e Telégrafos (ECT) tem se engajado no esforço de exploração do mercado brasileiro de mensagens h íbridas mediante a prestação de serviços de correio híbrido, os quais permitem a remessa eletrônica e a entrega física de correspondências. O objetivo do presente trabalho foi analisar o posicionamento estratégico mais adequado para a ECT no mencionado mercado. Para tanto, procurou-se atingir os seguintes objetivos intermediários: i) apresentar o conceito de correio híbrido, seu processo de funcionamento, prestadores e usuários potenciais do serviço; ii) descrever a ECT e analisar seu contexto de atuação; iii) avaliar o mercado potencial de mensagens híbridas que a ECT pretende explorar através dos serviços de correio híbrido e iv) analisar a estrutura da indústria postal no Brasil. Para a elaboração do estudo, adotou-se, como referencial teórico, os seguintes modelos conceituais desenvolvidos por Michael Porter (1989): as cinco forças competitivas, a cadeia de valores e as estratégias competitivas genéricas. A partir dos resultados obtidos, concluiu-se que a estrutura da indústria postal brasileira recomenda a adoção de uma posição de baixo custo e escopo competitivo amplo no mercado de mensagens híbridas. Além disso, verificou-se que as soluções de correio híbrido constituem um tipo de inovação que pode ter um enorme efeito positivo não apenas sobre o custo, mas também sobre a diferenciação. Deste modo, recomenda-se que a ECT explore todas as oportunidades de diferenciação abertas pelo correio híbrido que não comprometam a posição de baixo custo, visto ser este o alvo estratégico primário.

Electoral rules, political competition and fiscal spending: regression discontinuity evidence from brazilian municipalities

We exploit a discontinuity in Brazilian municipal election rules to investigate whether political competition has a causal impact on policy choices. In municipalities with less than 200,000 voters mayors are elected with a plurality of the vote. In municipalities with more than 200,000 voters a run-off election takes place among the top two candidates if neither achieves a majority of the votes. At a first stage, we show that the possibility of runoff increases political competition. At a second stage, we use the discontinuity as a source of exogenous variation to infer causality from political competition to fiscal policy. Our second stage results suggest that political competition induces more investment and less current spending, particularly personnel expenses. Furthermore, the impact of political competition is larger when incumbents can run for reelection, suggesting incentives matter insofar as incumbents can themselves remain in office.

A Common-feature approach for testing present-value restrictions with financial data

It is well known that cointegration between the level of two variables (labeled Yt and yt in this paper) is a necessary condition to assess the empirical validity of a present-value model (PV and PVM, respectively, hereafter) linking them. The work on cointegration has been so prevalent that it is often overlooked that another necessary condition for the PVM to hold is that the forecast error entailed by the model is orthogonal to the past. The basis of this result is the use of rational expectations in forecasting future values of variables in the PVM. If this condition fails, the present-value equation will not be valid, since it will contain an additional term capturing the (non-zero) conditional expected value of future error terms. Our article has a few novel contributions, but two stand out. First, in testing for PVMs, we advise to split the restrictions implied by PV relationships into orthogonality conditions (or reduced rank restrictions) before additional tests on the value of parameters. We show that PV relationships entail a weak-form common feature relationship as in Hecq, Palm, and Urbain (2006) and in Athanasopoulos, Guillén, Issler and Vahid (2011) and also a polynomial serial-correlation common feature relationship as in Cubadda and Hecq (2001), which represent restrictions on dynamic models which allow several tests for the existence of PV relationships to be used. Because these relationships occur mostly with nancial data, we propose tests based on generalized method of moment (GMM) estimates, where it is straightforward to propose robust tests in the presence of heteroskedasticity. We also propose a robust Wald test developed to investigate the presence of reduced rank models. Their performance is evaluated in a Monte-Carlo exercise. Second, in the context of asset pricing, we propose applying a permanent-transitory (PT) decomposition based on Beveridge and Nelson (1981), which focus on extracting the long-run component of asset prices, a key concept in modern nancial theory as discussed in Alvarez and Jermann (2005), Hansen and Scheinkman (2009), and Nieuwerburgh, Lustig, Verdelhan (2010). Here again we can exploit the results developed in the common cycle literature to easily extract permament and transitory components under both long and also short-run restrictions. The techniques discussed herein are applied to long span annual data on long- and short-term interest rates and on price and dividend for the U.S. economy. In both applications we do not reject the existence of a common cyclical feature vector linking these two series. Extracting the long-run component shows the usefulness of our approach and highlights the presence of asset-pricing bubbles.

Bank dividends and signaling to information-sensitive depositors

Esta tese investiga se a composição do endividamento dos bancos afeta sua política de dividendos. Identificou-se que investidores sensíveis a informações (investidores institucionais) são alvos de sinalização através de dividendos por parte dos bancos. Utilizando uma base de dados exclusiva de bancos brasileiros, foi possível identificar vários tipos de credores, especificamente, investidores institucionais, empresas não financeiras e pessoas físicas, que são alvos potenciais de sinalização por dividendos. Adicionalmente, a existência de vários bancos de capital fechado, controlados e geridos por um pequeno grupo de acionistas, em que a sinalização direcionada a acionistas é implausível, permite inferir que bancos que utilizam mais fundos de investidores sensíveis a informações (institucionais) pagam mais dividendos, controlando por diversas características. Durante a crise financeira, este comportamento foi ainda mais pronunciado. Esta relação reforça o papel dos dividendos como uma forma custosa e crível de comunicar sobre a qualidade dos ativos dos bancos. A hipótese de que os dividendos podem ser utilizados como uma forma de expropriação dos depositantes por parte dos acionistas é refutada, uma vez que, se fosse esse o caso, observar-se-ia esse maiores dividendos em bancos com depositantes menos sensíveis a informação. Além disso, foi verificada uma relação negativa entre o pagamento de dividendos e o custo de captação (juros pagos em certificados de depósito bancário) e uma relação positiva de dividendos com o tamanho e com os lucros passados, e que os bancos de capital fechado pagam mais dividendos do que os de capital aberto, uma descoberta que também se alinha com a ideia de que os depositantes seriam os alvos da sinalização por dividendos. Finalmente, encontrou-se também uma relação negativa entre dividendos e adequação de capital do bancos, o que indica que pressões regulatórias podem induzir os bancos a pagar menos dividendos e que o pagamento de dividendos é negativamente relacionado com o crescimento da carteira de crédito, o que é consistente com a ideia de que os bancos com maiores oportunidades de investimento retêm seus lucros para aumentar seu patrimônio líquido e sua capacidade de conceder crédito.

Are defaults a family affair?: the relationship between kinship and microloan performance

A partir de uma grande base de dados fornecida por uma instituição de microcrédito do nordeste brasileiro, a relação entre parentesco e inadimplência é analisada. O presente trabalho mostra evidências de que o parentesco entre membros de um grupo solidário afeta positivamente a adimplência. Grupos em que todos os membros possuíam algum laço de parentesco entre si apresentaram uma probabilidade 24% menor de inadimplência do que grupo sem esses laços. Além disso, quando considerado apenas o primeiro empréstimo, grupos com 100% de parentesco entre os membros apresentaram uma probabilidade 45% menor de inadimplência em comparação com membros sem relações de parentesco. Os resultados deste trabalho também mostram que há uma relação negativa entre parentesco e probabilidade de mudança na formação de um grupo solidário. Essas análises permitem fazer inferências acerca dos mecanismos pelos quais o parentesco afeta o desempenho do microcrédito. Os resultados sugerem que o parentesco possui um benefício maior no processo de autosseleção do grupo. Além disso, os resultados também sugerem que o impacto positivo do parentesco nas atividades de automonitoramento compensa o enfraquecimento das atividades de enforcement.

Generic-branded drug competition and the price for pharmaceuticals in procurement auctions

This paper studies the effects of generic drug’s entry on bidding behavior of drug suppliers in procurement auctions for pharmaceuticals, and the consequences on procurer’s price paid for drugs. Using an unique data set on procurement auctions for off-patent drugs organized by Brazilian public bodies, we surprisingly find no statistically difference between bids and prices paid for generic and branded drugs. On the other hand, some branded drug suppliers leave auctions in which there exists a supplier of generics, whereas the remaining ones lower their bidding price. These findings explain why we find that the presence of any supplier of generic drugs in a procurement auction reduces the price paid for pharmaceuticals by 7 percent. To overcome potential estimation bias due to generic’s entry endogeneity, we exploit variation in the number of days between drug’s patent expiration date and the tendering session. The two-stage estimations document the same pattern as the generalized least square estimations find. This evidence indicates that generic competition affects branded supplier’s behavior in public procurement auctions differently from other markets.

The implementation of corporate governance in Germany and Brazi l: a comparative case study

The recent promotion of best corporate governance standards by several different government institutions and non-for profit organizations resulted in the implementation of more sophisticated governance mechanisms. As consequence to the separation of ownership and control the concept of agency theory arose. Agency theory argues that without out proper control mechanism managers would behave exploit owners due to information asymmetry. Regulators have promoted corporate governance mechanisms in order to address this issue. This paper aims to contrast the implementation of best corporate governance practices in Germany and Brazil on the example of two practical examples. With this purpose in mind, this paper analyzed two companies listed in the main stock exchange in Germany and Brazil throughout a period of 5 years. In order to measure the degree of corporate governance practices implemented 3 different parameters have been chosen. In line with great part of the literature the parameters considered to be relevant are; composition, procedures and deviation from the local corporate governance code. The comparison of the data revealed that board composition in the two analyzed companies is similar regarding the proportion of independent representatives but does distinguish in size. While committees are related to the same topics it can be implied that Natura’s board is more involved in the actual management of the company. Lastly, Beiersdorf has been able to comply to a larger extend with the recommendations of the local German code than Natura to the recommendations published by Brazilian code of the IBGC.

Essays on banking theory and history of financial arrangements

Esta tese contém dois capítulos, cada um lidando com a teoria e a história dos bancos e arranjos financeiros. No capítulo 1, busca-se extender uma economia Diamond-Dybvig com monitoramento imperfeito dos saques antecipados e realizar uma comparação do bem estar social em cada uma das alocações possíveis, como proposto em Presscott and Weinberg(2003). Esse monitoramento imperfeito é implementado a partir da comunicação indireta ( através de um meio de pagamento) entre os agentes e a máquina de depósitos e saques que é um agregado do setor produtivo e financeiro. A extensão consiste em estudar alocações onde uma fração dos agentes pode explorar o monitoramento imperfeito e fraudar a alocação contratada ao consumirem mais cedo além do limite, usando múltiplos meios de pagamento. Com a punição limitada no período de consumo tardio, essa nova alocação pode ser chamada de uma alocação separadora em contraste com as alocações agregadoras onde o agente com habilidade de fraudar é bloqueado por um meio de pagamento imune a fraude, mas custoso, ou por receber consumo futuro suficiente para tornar a fraude desinteressante. A comparação de bem estar na gama de parâmetros escolhida mostra que as alocações separadoras são ótimas para as economias com menor dotação e as agregadoras para as de nível intermediário e as ricas. O capítulo termina com um possível contexto histórico para o modelo, o qual se conecta com a narrativa histórica encontrada no capítulo 2. No capítulo 2 são exploradas as propriedade quantitativas de um sistema de previsão antecedente para crises financeiras, com as váriaveis sendo escolhidas a partir de um arcabouço de ``boom and bust'' descrito mais detalhadamente no apêndice 1. As principais variáveis são: o crescimento real nos preços de imóveis e ações, o diferencial entre os juros dos títulos governamentais de 10 anos e a taxa de 3 meses no mercado inter-bancário e o crescimento nos ativos totais do setor bancário. Essas variáveis produzem uma taxa mais elevada de sinais corretos para as crises bancárias recentes (1984-2008) do que os sistemas de indicadores antecedentes comparáveis. Levar em conta um risco de base crescente ( devido à tendência de acumulação de distorções no sistema de preços relativos em expansões anteriores) também provê informação e eleva o número de sinais corretos em países que não passaram por uma expansão creditícia e nos preços de ativos tão vigorosa.