New publicationHow data structure and dependencies influence statistics
27 June 2025

Photo: privat
Dr. Zihao Yuan and Dr. Prof. Dr. Martin Spindler publish their research results in the renowned Journal of Econometrics.
The article “Bernstein-type Inequalities and Nonparametric Estimation under Near-Epoch Dependence” by Dr. Zihao Yuan and Prof. Dr. Martin Spindler has been accepted for publication at the Journal of Econometrics. Dr. Yuan was a PhD student at the professorship of statistics until end of 2024 and is now a postdoc with Prof. Dette at the Ruhr-University Bochum. There, he conducts research in the field of “mathematical statistics”.
The paper makes two main contributions:
- It develops new mathematical inequalities for irregularly distributed data and shows that their goodness depends on the so-called effective dimension of the data distribution, which is introduced in the article.
- These inequalities are then used to investigate when certain estimation methods produce particularly good results – especially for data with spatial or temporal dependence (and infill asymptotics).
Overall, the work improves the understanding of how data structure and dependencies influence statistical methods.