Research

http://emap.fgv.br/linhasdepesquisa#934Data ScienceEduardo Fonseca Mendes, Flávio Codeço Coelho, Jorge Poco, Moacyr Alvim Horta Barbosa da Silva, Renato Rocha Souza, Rodrigo dos Santos Targino, Yuri Fahham Saporito
The objective of this line of research is the exploration and development of objects and research problems through methods originated from the statistical learning of machines (Machine Learning) and its applications in the analysis of structured and unstructured data. This range of empirical objects encompasses numerical, categorical and textual data, images, graphs, time series, among others. It also includes the development of algorithms, languages and methodologies for manipulating these complex databases, in tasks such as classification, regression, learning, identification of clusters, recommendation systems, optimization, extraction and representation of knowledge, modeling knowledge domains, data mining, text mining, visualization, sentiment analysis, highdimensional data analysis, cryptography.
Publicações recentes

http://emap.fgv.br/linhasdepesquisa#935Control and OptimizationMaría Soledad Aronna, Roger Behling, Vincent Gérard Yannick Guigues
The Optimal Control Theory studies Optimization problems whose state variables are subject to differential equations (ordinary or partial) whose dynamics depend on a control variable, while Stochastic Optimization is focused on the study of optimization problems involving uncertainties modeled by a stochastic process. A stochastic optimization problem is usually formulated using conditional risk measures.
Publicações recentes

http://emap.fgv.br/linhasdepesquisa#936EpidemiologyClaudio José Struchiner, Eduardo Massad, Flávio Codeço Coelho, Luiz Max Fagundes de Carvalho, María Soledad Aronna, Moacyr Alvim Horta Barbosa da Silva
Mathematical epidemiology is an applied mathematics subject with more than a century of tradition, which incorporates ideas and methods from different branches of mathematics in the representation and analysis of its objects of study.
At EMAp, the Mathematical Epidemiology group studies how information on how communicable diseases can be treated from a quantitative point of view, differently from the classic statistical techniques well known in the medical field. In addition, the projects require a basic understanding of biological assumptions and the mathematical models that underlie infection models and the tools available today to extract biological information from these models.Publicações recentes

http://emap.fgv.br/linhasdepesquisa#944StatisticsClaudio José Struchiner, Eduardo Fonseca Mendes, Flávio Codeço Coelho, Luiz Max Fagundes de Carvalho, Rodrigo dos Santos Targino, Vincent Gérard Yannick Guigues
In general, this line of research is divided into theoretical, methodological and applied aspects involving data analysis. In particular, EMAp researchers are interested in the following areas of Statistics: Estimation of time series in high dimension; Modeling of nonlinear time series; Nonparametric tests; Monte Carlo simulation for Bayesian inference; Statistical inference for riskaverse or neutral risk stochastic optimization problems (central limit theorems, hypothesis tests, nonasymptotic confidence intervals); Nonparametric hypothesis testing using convex optimization techniques.
Publicações recentes

http://emap.fgv.br/linhasdepesquisa#945Finance and RiskEduardo Fonseca Mendes, Flávio Codeço Coelho, Hugo A. de la Cruz Cancino, Rodrigo dos Santos Targino, Vincent Gérard Yannick Guigues, Yuri Fahham Saporito
In this line of research, mathematical models and methods are studied with a focus on applications in problems of Finance, Economics and Actuarial Sciences. For applications in Finance and Economics, research projects are usually based on techniques involving Stochastic Calculation and/or Partial Differential Equations, whereas applications in Actuarial Sciences, such as Reserves calculation, tend to use statistical regression models or time series.
Publicações recentes

http://emap.fgv.br/linhasdepesquisa#946Stochastic ModelsHugo A. de la Cruz Cancino, Yuri Fahham Saporito
This line of research began worldwide with the development of Probability and the formal construction of the Brownian Movement in the early 20th century. This is probably one of the areas of mathematics with the most applications in the real world. From a theoretical point of view, we are interested in the study of Stochastic Differential Equations and the incorporation of pathdependence in several aspects of the theory.
In addition, another aspect of this line is the numerical resolution of differential equations, in particular random (RDEs) and stochastic (EDEs). The theory of EDEs and RDEs are themes, at the crossroads of differential equations and stochastic processes, with a wide variety of applications in the modeling of phenomena and practical situations in which the quantities of interest are subject to random disturbances. Since obtaining solutions to these equations is rarely possible, much attention is paid to the construction and study of approximation methods, with good qualitative properties, for their integration and simulation.
Publicações recentes