ZüKoSt: Seminar on Applied Statistics

Would you like to be notified about these presentations via e-mail? Please subscribe here.

Modal title

Modal content

Spring Semester 2019

Date / Time

Speaker

Title

Location

15 March 2019
15:15-16:00

Benjamin Stucky
Pharma UZH

Details

ZüKoSt Zürcher Kolloquium über Statistik

Title	Biomarkers for Sleep Deprivation-Induced Impairments in Human Cognition
Speaker, Affiliation	Benjamin Stucky, Pharma UZH
Date, Time	15 March 2019, 15:15-16:00
Location	HG G 19.1
Abstract	The phenomena of sleep and cognition involve complex phenotype-genotype associations, i.e., complex relationships between observable traits and the genetic variation that contributes to the expression of those traits. There is a general belief that investigating such relationships requires large sample sizes. However, sleep- and cognition-related phenotype-genotype associations may be strengthened through carefully controlled laboratory studies that amplify a given cognitive phenotype by perturbing the organism through sleep deprivation and pharmacological interventions. Here we will discuss the contributions of some recent modifications to the LASSO method (Tibshirani, 1996) in identifying a gene expression biomarker panel, i.e. a set of readily measurable genetic indicators of cognitive impairment due to sleep deprivation. We illustrate these approaches on the basis of a RNA expression data set introduced by Uyhelji et al., 2018.

Biomarkers for Sleep Deprivation-Induced Impairments in Human Cognitionread_more

HG G 19.1

22 March 2019
15:15-16:00

Mollie Brooks
National Institute of Aquatic Resources, Technical University of Denmark

Details

ZüKoSt Zürcher Kolloquium über Statistik

Title	Flexible generalized linear models with glmmTMB
Speaker, Affiliation	Mollie Brooks, National Institute of Aquatic Resources, Technical University of Denmark
Date, Time	22 March 2019, 15:15-16:00
Location	HG G 19.1
Abstract	The diversity of features of generalized linear models that can be fit in R is huge (e.g. zero-inflation, numerous distributions, and random effect correlation structures), but the features were not easy to use in combination because they were available from separate packages. Ignoring these aspects can give biased estimates and inflate the rates of false-positives or false-negatives in hypothesis tests. The R package glmmTMB was developed using the TMB package to do maximum likelihood estimation to fit a diversity of models in a single robust package. Features of TMB that made this possible include automatic differentiation for calculating gradients, Laplace approximation for integrating over random effects, and adding and subtracting on the log-scale to avoid over- and under-flow. In addition to describing glmmTMB, the talk will include ecological examples that address zero-inflation and underdispersion in count data, as well as Tweedie and beta regression for biomass and proportion data, respectively.

Flexible generalized linear models with glmmTMBread_more

HG G 19.1

29 March 2019
15:15-16:00

Sebastian Sippel
Institute for Atmospheric and Climate Science, ETH

Details

ZüKoSt Zürcher Kolloquium über Statistik

Title	Climate change detection: Uncovering circulation-driven and external components in climate observations and models
Speaker, Affiliation	Sebastian Sippel, Institute for Atmospheric and Climate Science, ETH
Date, Time	29 March 2019, 15:15-16:00
Location	HG G 19.1
Abstract	Internal atmospheric variability fundamentally constrains short- and medium-term climate predictability and obscures detection of anthropogenic climate change on regional scales. Dynamical adjustment is a traditional climate science technique to characterize circulation-induced variability in temperature or precipitation; the residual contains an estimate of the external “forced”, i.e. circulation-independent response. Here, we present a novel dynamical adjustment technique that makes use of statistical learning principles within the context of (1) a set of climate model simulations and (2) a real-world application of the method to variability in winter temperatures and snow coverage in Switzerland. The statistical learning methods establish a consistent relationship between internal circulation variability and atmospheric target variables on a daily time scale with around 80% variance explained in European monthly winter temperature and precipitation; and to a similar degree for global annual mean temperature and zonal mean precipitation. A real-world application to the Swiss winter temperature series, a long-term homogenized observational record reveals that a large fraction of winter temperature variability, and to a smaller degree variability in snow coverage, can be explained by internal atmospheric variability. The adjusted residual time series reveals a smooth, increasing trend since around the late 1970s, mostly driven by thermodynamic changes. Overall, statistical learning techniques for dynamical adjustment help to uncover the external forced response at regional and global scales, thus strengthening process understanding and facilitating detection of climate change.

Climate change detection: Uncovering circulation-driven and external components in climate observations and models read_more

HG G 19.1

5 April 2019
15:15-16:00

Ulrik Brandes
ETH, Departement Geistes-, Sozial- und Staatswissenschaften

Details

ZüKoSt Zürcher Kolloquium über Statistik

Title	Network science as the dual of statistics
Speaker, Affiliation	Ulrik Brandes, ETH, Departement Geistes-, Sozial- und Staatswissenschaften
Date, Time	5 April 2019, 15:15-16:00
Location	HG G 19.1
Abstract	If statistics is concerned with the collection, management, analysis, presentation, and interpretation of data, then so is network science. The distinction lies neither in a universal network theory nor the complex systems of certain application domains but (i) an incidence structure relating the units of observation and (ii) a corresponding focus on dependencies among observations. While not mainstream, this view clears away some of the smoke and mirrors obscuring the close relationship between statistics and network science. Following this rather principled line of thought, I will introduce the positional approach to network analysis as a means to capitalize on it.

Network science as the dual of statisticsread_more

HG G 19.1

Notes: if you want you can subscribe to the iCal/ics Calender.

Archive: SS 25 AS 24 SS 24 AS 23 SS 23 AS 22 SS 22 AS 21 SS 20 AS 19 SS 19 AS 18 SS 18 AS 17 SS 17 AS 16 SS 16 AS 15 SS 15 AS 14 SS 14 AS 13 SS 13 AS 12 SS 12 AS 11 SS 11 AS 10 SS 10 AS 09