The lecture will present the counting process point of view of the survival analysis. The mathematical tools of counting processes should not be considered only as pure technicalities, but they should be perceived as important theoretical concepts and it is neccessary to understand them. The counting process approach to the survival analysis enables significant generalizations to many situations in event history analysis (e.g. multistate models, recurrent event data). In addition to the brief introduction to the mathematical theory, a derivation of the Nelson-Aalen estimator, an alternative to the Kaplan-Meier estimator of the survival curve, will be given.
Quality of Health Care in Slovenia
Photos: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10.
Use of statistics in the field of quantitative geneticsQuantitative genetics tries to answer how much of the phenotypic variability is due to genetic variability. Statistical methods play an essential role in this task and some of these methods (e.g., regression, analysis of variance, mixed model, threshold model, ...) were developed in conjuction with the methodological development in quantitative genetics. These methods will be presented from the point of view of statistics and quantitative genetics and their use in the field of animal breeding and genetics.
Photos: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10.
Statistical issues emerging in training and evaluating classification models in presence of rare eventsThe problem of modeling binary responses by using cross sectional data has found a number of satisfying solutions extending throughout both parametric and nonparametric methods. Examples are traditional classification models like logistic regression, discriminant analysis, classification trees or procedures at the forefront as neural networks or combinations of classifiers (bagging, boosting, random forests). These models are based on the implicit assumption that the distribution of the responses is well balanced over the sample. However, there exist many real situations where it is a priori known that one of the two responses (usually the most interesting for the analysis) is rare. This class imbalance occurs in several domains as for example finance (detection of defaulter credit applicants), epidemiology (diagnosis of rare diseases), social sciences (analysis of anomalous behaviors), computer sciences (identification of some features of interest in image data). The class imbalance heavily afiects both the model estimation and the evaluation of its accuracy. Classification methods are in fact conceived to estimate the model that best fits the data according to some criterion of global accuracy. When data are unbalanced the model tends to focus on the prevalent class and ignore the rare events (Japkowicz, and Stephen, 2002). Moreover, when evaluating the quality of the classification, the same measures of global accuracy may lead to misleading results or even if alternative error measures are used, the scarcity of data conducts to high variance estimates of the error rate, especially for the rare class. In this work an unified and systematic framework for dealing with both the problems is proposed, based on a smoothed bootstrap form of re-sampling from data. The proposed technique includes some of the existing solutions as a special case, it is supported by a theoretical framework and reduces the risk of model overfitting. The presented talk is based on joint work with prof. Nicola Torelli from University of Trieste.
Slides: PDF
Photos: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15.
About experiments
Slides: PDF
Geographic analysis in oncological epidemiologySlides: PDF
Measuring ego-centered social networksSlides: PDF
Measures of AgreementThe talk will provide an overview of various agreement measures and different approaches to measuring agreement:
Slides: PDF