Seminars/Colloquia

ODSL Forum: "The Needle in the Haystack"

by Philipp Eller (TUM)

Europe/Berlin
Description

Presenter: Philipp Eller, TUM
Title: The Needle in the Haystack

Abstract: Detecting small deviations of samples from a hypothesized distribution, or in other words, assessing the goodness-of-fit of a distribution given a number of random samples is an often-encountered problem in data analysis. Such statistical hypothesis tests find applications in many fields, ranging from the natural and social sciences over engineering to quality control.

In this talk, I will give a brief overview over existing tests, such as well known "Kolmogorov-Smirnov" test, and also lesser known ones such as "Moran" or "Greenwood". I will then introduce our newly developed statistical test based on the observed spacings of ordered data. This new statistic is sensitive to detect non-uniformity in random samples, or short-lived features in event time series. Under some conditions, this new test can outperform existing ones, in particular when the number of samples is small and differences occur over a small quantile of the null hypothesis distribution.

Reading material: Philipp Eller & Lolian Shtembari, "A goodness-of-fit test based on a recursive product of spacings",  https://arxiv.org/abs/2111.0225

As usual, the format will be a short presentation followed by plenty of discussion.

Please let all interested know about the Journal Club (get them to send their Email address).  We are looking forward to lively discussions.