You are here: Home / About / Research


  • Statistic

Research Areas

  • Machine Learning

Machine learning was born within artificial intelligence as a set of tools to assist computational tasks using databases. In the last two decades, this area has expanded its reach and started to be used in the most varied areas of human knwoledge. Currently, machine learning consists of two major areas: supervised learning (which consists of making predictions for future observations) and unsupervised learning (which consists of learning more about the structure of data). The PIPGEs has been working in this research line along the following topics: inference and development of complex networks, incorporation of selection bias in machine learning models, quantification, use of machine learning methods for statistical inference, applications to law, biology, physics and other areas of knowledge.

  • Statistical Inference

Given the central role of statistical inference, it relates to several specific sub-areas. Several of these sub-areas are of active interest in the PIPGEs. For example survival and reliability analysis, bayesian methods, latent variable models, time series, fundaments of statistics...

  • Regression Models

This line has an outstanding importance in the field of statistics and the PIPGEs has developed works on linear and nonlinear models (including mixed models), one topics such as longitudinal data analysis, survival and reliability analysis, models with variable errors, latent variables models, models for discrete data, models for limited range data, models for credit score, time series and item response theory. These researches are developed on both frequentist and Bayesian methods and include diagnostic methods to evaluate different aspects of the proposed models.

  • Probability and Stochastic Processes

This line of research evolved in recent years, in the program, with researchs on theoretical and applied topics of stochastic processes. Particularly researches on stochastic calculus and applications (finance, asset pricing and hedging), particle systems and percolation; stochastic modeling of complex systems; asymptotic theorems in special stochastic processes; simulation algorithms and non-Markovian models.