Speaker: Riccardo De Bin (Department of Mathematics,UiO)
Title: Strategies to handle mandatory covariates using model- and likelihood-based boosting
Abstract: Among the iterative methods exploited during recent years in statistical practice, particular attention has been focused on boosting. Originally developed in the machine learning community to handle classification problems, boosting has been successfully translated into the statistical field and extended to many statistical problems, including regression and survival analysis. In a parametric framework, the basic idea of boosting is to provide estimates of the parameters by updating their values iteratively: at each step, a weak estimator is fitted on a modified version of the data, with the goal of minimizing a loss function. Thanks to its resistance to overfitting, boosting is particularly useful in the construction of prediction models. Its iterative nature, moreover, allows straightforward adaptations to cope with high-dimensional data. In this talk, we first review and contrast two well-known boosting techniques, model-based boosting and likelihood-based boosting. We note that in the simple linear regression case they lead to the same results, provided there is a specific choice for their tuning parameters. This is not the case for more complex situations. As an example, we show the differences in survival analysis under the proportional hazards assumption. As a main contribution of the talk, we analyze strategies to include mandatory variables, i.e. those variables that for some reasons must enter in the final model, in a statistical model using the two boosting techniques. In particular, we examine solutions currently only considered for one and explore the possibility of extending them to the other. We show the importance of a good handling of mandatory variables in a real data example.
Dear All in BigInsight. Hope you are having an exciting start of 2017!
I wish to inform you about a few changes in our organization.
Håvard Rue has moved from NTNU to take a professor position in Saudi Arabia, and therefore leaves BigInsight as codirector. We thank him very much for his help so far and wish him success in his new exciting position.
We welcome Associate Professor Ingrid Hobæk Haff, from the Department of Mathematics of UiO and Assistant Research Director PhD Anders Løland, from Norsk Regnesentral, as two new co-directors of BigInsight!
Ingrid Hobæk Haff takes also the responsibility of co-PI of the Innovation Objective area Fraud, which was also left free by the departure of Håvard.
We thank Professor Arne Bang Huseby for his help in the first years as co-PI of the IO Power, and welcome Professor Carlo Mannino, from the Department of Mathematics of UiO, as new co-Pi of the the Innovation Objective area Power.