Servicenavigation

Statistik: mehr als Erbsen zählen

Sie sind hier:

Meldungen-Archiv

SFB 876 Vortrag

Title: Reliable Data Mining in Uncertain Data

Speaker: Andreas Zuefle (George Mason University, USA)

When: 2020-01-30 16:15

Where: OH 14, E023

Our ability to extract knowledge from data is often impaired by unreliable, erroneous, obsolete, imprecise, sparse, and noisy data. Existing solutions for data mining often assume that all data are uniformly reliable and representative. Oblivious to sample size and sample variance, it is clear that mined patterns may be spurious, that is, caused by random variations rather than a causal signal. This is particularly problematic if latent features and deep learning methods are used to mine patterns, as their lack of interpretability prevents domain experts and decision makers from explaining spurious conclusions. This presentation will survey data mining algorithms that can exploit reliability information of data to enrich mined patterns with significance information. In detail, we will discuss the use of Monte Carlo and agent-based simulation to gain insights on the reliability of data mining results and we will look at applications for handling.

 

Vortragsankündigung

Title: Strong Convergence of Reliability Estimators for Multiple-Component Measuring Instruments in the Behavioral, Educational and Social Sciences

Speaker: Prof. Tenko Raykov (Michigan State University)

When: Tuesday, December 17, 2019, 16:30

Where: M / E21

This talk shows that in general the popular coefficient alpha estimator for reliability of multicomponent measuring instruments converges almost surely to a quantity that is not equal to the population reliability coefficient. This convergence with probability 1 is a stronger statement than convergence in probability and convergence in distribution for the alpha estimator, which have been studied in the past. In the special case of congeneric measures with uncorrelated errors and equal loadings on the common true score, the alpha estimator converges almost surely to the population reliability coefficient that equals population alpha, which implies its consistency as a reliability estimator. When the loadings are unequal but sufficiently high and similar, the alpha estimator converges also almost surely to population alpha that is essentially indistinguishable from the reliability coefficient, which implies alpha’s approximate consistency then. For the general case, the results entail that the alpha estimator is not a consistent estimator of reliability. The findings add (i) to the critical literature on coefficient alpha in the general case, as well as (ii) to the justification of its use as a dependable measuring instrument reliability estimator in special cases and settings resulting under appropriate restrictive conditions. The discussion is illustrated using a numerical example.

 

Statistik im Beruf

28.11.2019 Arne Ring (Head of Biometrics and Statistical Programming, medac GmbH, Wedel)

Titel:  "Pendler zwischen Pharma-Industrie und Universität"

Prof. Dr. Arne Ring hatte als Aktuar in der Rückversicherung begonnen, und war dann in die Pharma-Industrie gewechselt. Nach zwei Stationen an englischen Universitäten arbeitet er derzeit bei der medac in Wedel als Head of Biometrics and Statistical Programming. Regelmäßig betreut er eine Reihe von Masterarbeiten, insbesondere mit dem African Institute of Mathematical Sciences (AIMS), und hat dabei verschiedene afrikanische Staaten besucht. EIn Erfahrungsbericht findet man unter:

www.egms.de/static/de/journals/mibe/2019-15/mibe000195.shtml .

Die aktuelle Anzeige für Praktikanten und Abschlussarbeiten findet sich unter:

www.medac.de/karriere/alle-freien-jobs/jobs/hochschulpraktikant-student-intern/andere/661/

Der Vortrag findet am 28.11.2019 um 16.15 Uhr in Raum M/E 21 statt.

 

 

Tag der offenen Tür

Physik & Musik

Samstag 23. November, 10:30 Uhr Prof. Dr. Claus Weihs

Vom Klang zu Noten und Genres - Was Musik mit Statistik zu tun hat

 

Statistik im Beruf

07.11.2019 Tim Søren Beige (Statistical Analyst, Lowell Finances Service GmbH, Ratingen)

Titel: "Vom Studenten zum Statistiker im Forderungsmanagement der Lowell Group"

Vor einem Jahr hat die Kooperation der Lowell Group mit der TU Dortmund begonnen und ist seit diesem Semester offiziell. Der Vortrag von Tim Søren Beige umfasst daher erstens seinen persönlichen Weg vom Studenten zum Statistical Analyst im Forderungsmanagement der Lowell Group. Zweitens gibt er aber auch Aufschluss über die aktuellen Angebote für Studierende aus statistisch-mathematischen Studiengängen sowie weitere aktuelle Kooperationsprojekte mit der TU Dortmund.

Der Vortrag findet am 07.11.2019 um 16.15 Uhr in Raum M/E 21 statt.

 

Invitation AG DANK autumn workshop

We would like to invite you to the AG DANK autumn workshop.

For further information and the preliminary program please refer to the homepage http://www.gfkl.org/ag-dank-herbsttagung-2019/

If you have any questions or remarks don’t hesitate to contact us.

For registration, please contact us.

We are looking forward to see you in Dortmund next month.

Karsten Lübke und Claus Weihs

E-Mail: karsten.luebke@fom.de, claus.weihs@tu-dortmund.de

 

 

SFB 823 Vortrag

Rethinking control chart design and evaluation

Speaker: William H. Woodall (Department of Statistics, Virginia Polytechnic Institute and State University, USA)

When and where?

Tuesday, October 8, 2019, 4.30 pm, M / E21