CV Lattes
ResearcherID
ORCID
Google Scholar Citations
Universidade Estadual de Campinas (UNICAMP). Instituto de Matemática, Estatística e Computação Científica (IMECC) (Institutional affiliation from the last research proposal) Birthplace: Brazil
Holds a PhD in Statistics from the University of North carolina at Chapel (1997), a MS in Statistics from University of Campinas (1992) and a BS in Statistics from ENCE-IBGE (1989). He is a Full Professor of Statistics at University of Campinas. His research focus on Statistics, specially Non-Parametric Statistics in the following subjects: U-statistics, wavelets, time series analysis and genetics. (Source: Lattes Curriculum)
| News published in Agência FAPESP Newsletter about the researcher |
| More itemsLess items |
| TITULO |
| Articles published in other media outlets ( ): |
| More itemsLess items |
| VEICULO: TITULO (DATA) |
| VEICULO: TITULO (DATA) |
We consider here heteroskedastic time series models, specifically from the family $GARCH(p,q)$, both on time domain and well as on the frequency domain. For the former we estimate the $(p+q)$-dimensional parameter through $M$-quantiles. On the latter, we have two situations. First we employ the discrete Fourier transform (DFT) to estimate the spectrum. Further, we discuss the cpestrum, d…
This project is a natural follow-up to an ongoing project 18/04654-9, which is the most recent of a series of thematic projects from this research group endowed by FAPESP in the last twenty years. The proposed statistical methodologies will have application in Medicine, Finance, Biology, Physics, Neurosciences, Engineering, among others areas. They solve theoretical and empirical question…
(Only some records are available in English at this moment)
Clustering and classification methods are routinely used in machine learning. This application on a massive scale of techniques devised for low dimensional data may yield mediocre results. These scalability issues are true for iid data, time series and images. Some of the classical methods include K-th nearet neighbors, K-means, support vector machines, trees and florests, Schur measures,…
Spatial covariance models have in general dense matrices. Therefore, estimation and prediction depend on approximations. This problem is a major concern in hierarchical models, such as the Poisson regression. We are going to study new approaches of approximate inference based upon Pólya-Gamma variables augmented likelihood via Hamiltonian Monte Carlo. We aim to provide concrete performanc…
This project proposes new statistical approaches to the problem of variable selection in functional regression models using the Bayesian paradigm. Although frequentist methodologies are well understood, using Bayesian inference in functional data has still not been explored. Variational inference techniques, widely applied in Statistical Machine Learning, demonstrate great efficiency, but…
(Only some records are available in English at this moment)
Big data and artificial intelligence brings forward several notable technological advancements for decades to come. Assessing relevant information from large volume of data depends on fast and reliable algorithms. Two exampled here ar feature screening and statistical methods for the dynamics of serial data. The first task consists in discriminating which variables are noise, and which ha…
Monitoring the environment is an increasingly relevant and demanding area of human knowledge. For this, the development of methods and tools which are at the same time reliable, feasible and escalable are paramount. This project proposes some statistical measures for the characterization of changes on time series of SAR images. The paradigm of such measures is the multi-atribute represent…
We consider a high dimensional dynamic regression model. This means that we have a high number of covariates as well as a large number of sampled points on time, with a corresponding substantial dynamic effect. Some basic screening techniques are studied, and the efficiency of pre-screening dimensional reduction is discussed as well. Both paradigms of generalized correlation ands dimensio…
(Only some records are available in English at this moment)
Change detection is an important task performed in remote sensing image that allows researchers and engineers to identify and evaluate modificationson land surfaces captured by multi-temporal satellite images. Analyzing problems such as deforestation (Barreto et al., 2016), rapid urbanization (Ban and Yousif, 2012) and glacier melting (Scher, Steiner, and McDonald, 2021) are of great impo…
Wavelets methods have been applied on many fields of science. However, many relevant applications involve non-usual domains, as is the case with graph spaces. Moreover, due to improvement of data acquisition and storage, data can be sometimes collected along the time, allowing to evaluate the progress of the analyzed phenomenon as a time series. Therefore, the development of statistical m…
The theory of wavelets and Multiscale analysis motivate a large proportion of modern statistical methodology and advances in non-statistical fields have led to important, even indispensable statistical tools. Massive data sets, functional data, and high-frequency sampled processes intrinsically invariant to changes in scale are increasingly observed and stored. As a result, it is often cl…
(Only some records are available in English at this moment)
| 3 / 3 | Ongoing grants |
| 15 / 12 | Completed research grants |
| 4 / 4 | Ongoing scholarships in Brazil |
| 15 / 12 | Completed scholarships in Brazil |
| 6 / 6 | Completed scholarships abroad |
| 43 / 37 | All research grants and scholarships |
| Associated processes | |