Líneas de Investigación

The research areas may be develop by our students are:
  • Biostatistics: considers both methodological development and applications motivated by problems of biology and medicine, including genetics, clinical trials, and longitudinal studies.
  • Statistical Computing: Studies efficient algorithms for the resolution of problems originated in the application of statistical methods. These include maximum likelihood or Bayesian estimation (parametric and non-parametric) for item response models, calculation of ROC curves, models for longitudinal data and non-linear regression with multiple types of covariates, among others. The study includes both the efficient computational implementation of the methodology as well as some of its theoretical properties.

  • Bayesian Methods: considers the development and study of probability model properties from a Bayesian point of view, both parametric and non-parametric. It also includes aspects related to the development of efficient methods to implement statistical inference through a posteriori simulation.

  • Sampling: study and develop sampling designs for the efficient collection of data from a target population. Methods are developed to make statistical inference according to the designs in question. Applications are made in several areas such as geology, ecology, epidemiology and social sciences.

  • Psicometry and Educational Measurement : focused on the statistical modeling of problems motivated by the Chilean educational policy. Some of the topics addressed by the line are: item response theory, added value and school effectiveness, scoring and equating, problems of identification in item response theory models.

  • Time Series and Finantial Applications: considers the development of statistical methods for the analysis of time series, as well as the study and prediction of financial series that result from the evolution of economic indexes and financial market instruments.

  • Distribution Theory: i) Asymmetric distributions, important for generalizing linear elliptic models, ii) Sensitivity analysis to study the effect of perturbations in the assumptions of the model or modifications in the data, iii) Errors in the Variables, frequently present in diverse areas of knowledge, iv) Identifiability in structural models with emphasis on psychometrics.