Dr. rer. nat. Jonathan Ansari
Leading the research project ‚SORT: Stochastic orders for functional dependence‚.
Contact: jonathan.ansari(at)plus.ac.at
Research Interests
- Dependence Modeling
- Optimal Transport
- Multivariate and Nonparametric Statistics
- Stochastic Orderings
- Risk Analysis
- Financial Mathematics
- Data Analytics and Machine Learning
News
- New preprint: In my paper Copulas farthest from independence in quadratic Wasserstein distance, I solve the extremal problem of identifying copulas that are farthest from independence in quadratic Wasserstein distance and determine the maximal distance explicitly. The proof combines optimal transport with the conditional convex order – an ordering for the strength of functional dependence, recently introduced in Ansari & Fuchs (2025).
- New preprint: In joint work with M. Rockel and S. Steinmaßl, we provide The exact region determined by Spearman’s rho and Gini’s gamma over the class of bivariate copulas. In particular, we derive sharp bounds between the two dependence measures and characterize the copulas attaining the boundary.
- New preprint: In work On a copula product characterizing independence and perfect functional dependence, I introduce a new copula product that provides a unified characterization of independence and perfect functional dependence. The construction connects Wasserstein-based dependence measures with increasing rearrangements and reveals new structural links between copula products and measures of functional dependence.
- New preprint: In joint work with Johannes Wiesel from University of Copenhagen, our paper Dependence Measures via Adapted Optimal Transport: Stability and Rates of Convergence is available on arXiv. Core contributions: Standard dependence measures—such as Chatterjee’s rank correlation—lack weak continuity, which limits the applicability of empirical estimators. We address this by introducing a conditional continuity framework that restores convergence properties. We determine O(N^{-1/3}) rates of convergence for plug-in estimators, providing a robust statistical foundation for a broad class of rearranged and rank-based measures.
- New publication: In joint work with Marcus Rockel from University of Freiburg, our paper on The exact region and an inequality between Chatterjee’s and Spearman’s rank correlations has been published in the Journal of Multivariate Analysis.
- New publication: In joint work with Eva Lütkebohmert from University of Freiburg, our paper on Robust Bernoulli Mixture Models for Credit Portfolio Risk has been published in Mathematical Finance.bst Bernoulli Mixture Models for Credit Portfolio Risk
- My FWF research proposal on ‘Stochastic orders for functional dependence (SORT)’ with a funding amount of over 450,000€ and a duration of 3 years was approved (Grant-DOI 10.55776/PAT1669224)
- New preprint: In joint work with Sebastian Fuchs, our paper ‚On continuity of Chatterjee’s rank correlation and related dependence measures is available‚ on arXiv.
- In Feb. 2024, I won an Early Career Research grant from the Paris Lodron University Salzburg. My research project on improved risk bounds for sums of random variables under dependence information is being funded with over €50,000.