Searching for New Phenomena with Likelihood Ratio Tests: An Overview of Non-Standard Scenarios

The likelihood ratio test is a standard statistical tool widely used to perform tests of hypotheses. The null distribution of the likelihood ratio test statistic is often assumed to be χ2, following Wilks’ theorem. However, in many circumstances relevant to modern physics experiments, this theorem is not applicable. Examples include detecting signals whose location is unknown, distinguishing new sources from known astrophysics mimicking them, and searching for new physics under background mismodeling. In this talk, I will overview practical ways to identify situations of non-regularity and I will discuss solutions to construct valid inference in these settings.

Please join the event.

About Sara Algeri

I am an Assistant Professor in the School of Statistics at the University of Minnesota. My research interests mainly lie in astrostatistics, statistical inference, and computational statistics. The main purpose of my work is to provide highly generalizable statistical solutions that directly address fundamental questions in the physical sciences, and can at the same time be easily applied to any other scientific problem following a similar statistical paradigm. In line with this, motivated by problems arising in high-energy physics and astronomy, my current research focuses on statistical inference for signal detection, background estimation, and model assessment.