{"ID":2838897,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.16029","arxiv_id":"2511.16029","title":"Possibilistic Instrumental Variable Regression","abstract":"Instrumental variable regression is a common approach for causal inference in the presence of unobserved confounding. However, identifying valid instruments is often difficult in practice. In this paper, we propose a novel method based on possibility theory that performs posterior inference on the treatment effect, conditional on a user-specified set of potential violations of the exogeneity assumption. Our method can provide informative results even when only a single, potentially invalid, instrument is available, offering a natural and principled framework for sensitivity analysis. Simulation experiments and a real-data application indicate strong performance of the proposed approach.","short_abstract":"Instrumental variable regression is a common approach for causal inference in the presence of unobserved confounding. However, identifying valid instruments is often difficult in practice. In this paper, we propose a novel method based on possibility theory that performs posterior inference on the treatment effect, con...","url_abs":"https://arxiv.org/abs/2511.16029","url_pdf":"https://arxiv.org/pdf/2511.16029v2","authors":"[\"Gregor Steiner\",\"Jeremie Houssineau\",\"Mark F. J. Steel\"]","published":"2025-11-20T04:20:42Z","proceeding":"stat.ME","tasks":"[\"stat.ME\",\"econ.EM\",\"math.ST\"]","methods":"[]","has_code":false}