Seminar overview
×
Modal title
Modal content
Autumn Semester 2015
Date & Time | Speaker | Title | Location |
---|---|---|---|
Fri 18.09.2015 15:15-16:00 |
Robin Evans University of Oxford |
Abstract
Directed acyclic graph models (DAG models, also called Bayesian networks) are widely used in the context of causal inference, and they can be manipulated to represent the consequences of intervention in a causal system.
However, DAGs cannot fully represent causal models with confounding; other classes of graphs, such as ancestral graphs and ADMGs, have been introduced to deal with this using additional kinds of edge, but we show that these are not sufficiently rich to capture the range of possible models. In fact, no mixed graph over the observed variables is rich enough, however many edges are used.
Instead we introduce mDAGs, a class of hyper-graphs which is appropriate for representing causal models when some of the variables are unobserved.
Results on the Markov equivalence of these models show that when interpreted causally, mDAGs are the minimal class of graphs which can be sensibly used. Understanding such equivalences is critical for the use of automatic causal structure learning methods, a topic in which there is considerable interest. We elucidate the state of the art as well as some open problems.
Research Seminar in StatisticsCausal Models and How to Refute Themread_more |
HG G 19.1 |
Fri 25.09.2015 15:15-16:00 |
Damian Kozbur Seminar für Statistik |
Abstract
This work develops theory for a Testing-Based Forward Model Selection procedure in linear regression and m-estimation problems. Forward Selection is a greedy procedure which inductively selects covariates which reliably add predictive power into a statistical model. A testing procedure, whose properties resemble traditional statistical hypothesis testing, is assumed as a primitive. It is used as a tool for guiding the forward selection by deciding which variable to include next and when to stop including variables. This offers a flexible and data- driven way to estimate models and construct predictions in high dimensional statistical problems. Probabilistic bounds for prediction error and number of selected covariates are proved for the proposed Testing- Based Forward Selection. The regularity conditions assumed on the set of hypothesis tests are verified for several example problems. The per- formance of Testing-based Forward Selection is compared to Lasso and Post-Lasso in Simulation studies.
Research Seminar in StatisticsTesting-Based Forward Model Selectionread_more |
HG G 19.1 |
Mon 28.09.2015 11:00-12:00 |
Claudia Klüppelberg Technische Universität München |
Abstract
We consider a new max-linear structural equation model, where all random variables can be written as a max-linear function of their parents and noise terms. For the corresponding graph we assume that it is a DAG. We present basic probabilistic results of our model; in particular we characterise those max-linear random vectors, which originate in a max-linear structural equation model and are, hence, max-linear distributions on DAGs. We also determine the minimal DAG corresponding to the max-linear structural equation model. As an example we consider the structural equation model with max-stable noise variables, resulting in a max-stable max-linear DAG.
This is joint work with Nadine Gissibl.
Research Seminar in StatisticsMore information: https://stat.ethz.ch/events/research_seminarcall_made Extremes on directed acyclic graphsread_more |
HG G 19.1 |
Fri 23.10.2015 15:15-16:00 |
Alessandra Luati University of Bologna |
Abstract
The paper introduces the class of generalised linear models with Box-Cox link for the spectrum of a time series. The Box-Cox transformation of the spectral density is represented as a finite Fourier polynomial, with coefficients, that we term generalised cepstral coefficients, providing a complete characterisation of the properties of the random process. The link function depends on a power transformation parameter and encompasses the exponential model (logarithmic link), the autoregressive model (inverse link), and the moving average model (identity link). One of the merits of this model class is the possibility of nesting alternative spectral estimation methods under the same likelihood-based framework, so that the selection of a particular parametric spectrum amounts to estimating the transformation parameter. We also show that the generalised cepstral coefficients are a one to one function of the inverse partial autocorrelations of the process, which can be used to evaluate the mutual information between the past and the future of the process.
Research Seminar in StatisticsGeneralised Linear Cepstral Models for the Spectrum of a Time Seriesread_more |
HG G 19.1 |
Tue 27.10.2015 15:15-16:15 |
Po-Ling Loh University of Pennsylvania |
Abstract
Robust statistics provides a framework for quantifying the behavior of an estimator when data are subject to various imperfections which deviate from standard model assumptions. Such an understanding may aid in designing estimators that are robust to particular forms of contamination. In this talk, we will discuss two high-dimensional estimators suited for robust linear regression and precision matrix estimation. The form of the estimators is quite natural and the analysis follows from standard high-dimensional techniques; what is fascinating is the relationship between the results we derive and their analogs in low dimensions. In particular, we demonstrate that the same guiding principles for selecting estimators in classical robust statistics provide provably good behavior in high dimensions, as well.
Research Seminar in StatisticsMore information: https://www.math.ethz.ch/sfs/news-and-events/research-seminar.htmlcall_made New perspectives for robust/high-dimensional estimationread_more |
HG E 41 |
Thr 05.11.2015 16:15-17:00 |
Carsten Schmitz Winton Capital, London |
Abstract
Winton is a systematic asset manager and has been using computers and data to construct trading systems since 1997. Access to large amounts of data allows the testing of a large number of hypotheses which may sound good and empowering - but brings with it rather new challenges. In fact, only thorough statistical analysis can help find the path through the jungle of allegedly significant results.
ZüKoSt Zürcher Kolloquium über StatistikBig Data and Selection Bias in Financeread_more |
HG G 19.1 |
Fri 06.11.2015 15:15-16:00 |
Egil Ferkingstad NTNU Trondheim, Norway / University of Iceland |
Abstract
After briefly introducing the INLA approach for Bayesian computation, I will discuss two topics we're currently working on. Topic 1: A new copula-based correction (http://arxiv.org/abs/1503.07307) for difficult cases, allowing us to push the boundaries of applicability of INLA while adding minimal computational cost. Topic 2: Providing sensible, default, weakly informative priors. I will present the ideas behind Penalized Complexity priors (http://arxiv.org/abs/1403.4630), focusing on work in progress showing how the idea of distributing the overall variance between different model components gives a good strategy for specifying joint priors for hierarchical models.
Research Seminar in StatisticsMore information: https://www.math.ethz.ch/sfs/news-and-events/research-seminar.htmlcall_made Recent progress in the R-INLA project for Bayesian computation read_more |
HG G 19.1 |
Thr 19.11.2015 16:15-17:00 |
Martin Poschcall_made Medical University of Vienna |
Abstract
Adaptive designs play an increasingly important role in clinical drug
development. They use accumulating data of an ongoing trial to decide on
modifications of different design aspects without undermining the
validity and integrity of the trial. Several types of adaptations have
been considered as early stopping for futility or success, sample size
reassessment and change of population.
Particularly appealing applications are the use of adaptive designs in
combined phase II/III studies with treatment selection at interim and
adaptive enrichment designs where subgroups with a differential
treatment effect can be selected. We review the adaptive
design methodology for a single null hypothesis and discuss extensions
to multiple hypotheses testing problems. Finally, the application of
multi-stage designs in high dimensional testing problems is discussed.
ZüKoSt Zürcher Kolloquium über StatistikMore information: https://www.math.ethz.ch/sfs/news-and-events/seminar-applied-statistics.htmlcall_made Addressing Multiple Objectives in Clinical Trials: Adaptive Designs and Multiple Testingread_more |
HG G 19.1 |
Fri 04.12.2015 15:15-16:00 |
Benedikt Pötscher University of Vienna |
Abstract
Testing restrictions on regression coefficients in linear models often requires correcting the conventional F-test for potential heteroskedasticity or autocorrelation amongst the disturbances, leading to so-called heteroskedasticity and autocorrelation robust (or consistent) test procedures (HAR/HAC-tests). These procedures have been developed with the purpose of attenuating size distortions and power deficiencies present for the uncorrected F-test. We develop a general theory to establish positive as well as negative finite-sample results concerning the size and power properties of a large class of heteroskedasticity and autocorrelation robust tests. Using these results we show that nonparametrically as well as parametrically corrected F-type tests in time series regression models with stationary disturbances have either size equal to one or nuisance-infimal power equal to zero under very weak assumptions on the covariance model and under generic conditions on the design matrix. In addition we suggest an adjustment procedure based on artificial regressors. This adjustment resolves the problem in many cases in that the so-adjusted tests do not suffer from size distortions. At the same time their power function is bounded away from zero. As a second application we discuss the case of heteroskedastic disturbances.
Research Seminar in StatisticsFinite-Sample Properties of HAC-Testsread_more |
HG G 19.1 |
Thr 10.12.2015 16:15-17:00 |
Andrea Kraus Masaryk Unversity, Brno, Czech Republic |
Abstract
Epidemics have shaped our history, and even in modern times of huge scientic progress, major outbreaks are not uncommon. It is comparatively easy to tackle a small epidemic but very dicult to contain a large one, which urges public authorities to be on watch for early signs of new outbreaks. Mathematical and probabilistic modelling and statistical inference play a crucial role in these eorts. We give an overview of models for the spread of epidemics with focus on branching-process approximations for the initial stages. These originally population-growth models hinge on strong probabilistic concepts, such as the Markovian nature, rather than on specic information on the disease. This makes them suitable for use when the available information is largely conned to temporally aggregated counts of new cases subject to unknown amount of under-reporting.
Employing these models, we discuss the estimation of the spreading potential of an on-going epidemic in its early stages.
ZüKoSt Zürcher Kolloquium über StatistikMore information: https://www.math.ethz.ch/sfs/news-and-events/seminar-applied-statistics.htmlcall_made Modelling and estimating the spread of an epidemic from little initial informationread_more |
HG G 19.1 |
Fri 15.01.2016 15:15-16:00 |
Benjamin Frot Oxford University Statistics |
Abstract
We consider the problem of learning a conditional Gaussian graphical model in the presence of latent variables. Building on recent advances in this field, we suggest a method that decomposes the parameters of a conditional Markov random field into the sum of a sparse and a low-rank matrix. We derive convergence bounds for this estimator and show that it is well-behaved in the high-dimensional regime as well as "sparsistent" (i.e. capable of recovering the graph structure). We then describe a proximal gradient algorithm which is able to fit the model to thousands of variables. Through extensive simulations, we illustrate the conditions required for identifiability and show that there is a wide range of situations in which this model performs significantly better than its counterparts. We also show how this problem is relevant to some of the challenges faced by instrumental variable methods.
Research Seminar in StatisticsLatent variable model selection for Gaussian conditional random fieldsread_more |
HG G 19.1 |