26-27 May, 2019
To increase the development of human capital in computational statistics and support its progress, the Latin American Regional Section of the International Association of Statistical Computing (LARS-IASC) aims to implement a School on Computational Statistics and Data Science (LARS-IASC School). The main purpose of the School is to spread the knowledge base and advances in Computational Statistics in the Latin American countries and to increase the number of researchers and data scientists in the field.
The 2nd LARS-IASC School on Computational Statistics and Data Science will be a satellite event of ICORS-LACSC 2019. For its second edition, the LARS-IASC School will be on Robust Statistics. The main purpose is to provide an overview of the fundamental elements and recent advances in Robust Statistics. It is oriented to postgraduate students, researchers and practitioners in statistics. A solid knowledge on Statistical inference, Linear models, Multivariate Statistics and R programming is required from the participants. The program is organized around three axes: knowledge base and advances, software and data science applications.
The number of meeting hours are 12 hours. An official certificate will be given to participants at the end of the school. The number of participants is limited to 30 people on a first come first served basis.
- Introduction to robust statistics: concepts of influence function and breakdown value, univariate robust estimators (including explicit robust estimators of location, scale and skewness, as well as M-estimators of location and scale).
- Robust multivariate methods: Robust estimators of multivariate location and scatter (M-estimators, Stahel-Donoho, Minimum Covariance Determinant, S-estimators, MM-estimators)
- Robust regression: Robust linear regression (M-estimators, Least Trimmed Squares, S-estimators, MM-estimators, regression with categorical predictors)
- Robust PCA: principles, projection pursuit, spherical PCA, ROBPCA.
- Advanced topics and recent developments: inference (fast and robust bootstrap), multivariate and functional depth, high-dimensional data and sparsity, cellwise outliers.
International Organising Committee
- Holger Cevallos-Valdiviezo (Ecuador)
- Alba Martinez Ruiz (Chile)
- David F. Muñoz (Mexico)
- Paulo Canas Rodrigues (Brazil)
Registrations for 2nd LARS SCHOOL are open.
* Please consider that the number of registrations for 2nd LARS-IASC School will be limited to 30 people on a first come first served basis.