Carlos Cinelli (University of Washington)

Monday, September 19, 2022 - 3:40 pm to 5:00 pm
Event Type: 

Carlos Cinelli University of WashingtonDescription: Department Seminar: Carlos Cinelli (University of Washington)

Location: 368A Heady Hall

Contact Person: Otávio Bartalotti

Title: “Transparent and Robust Causal Inference in Econometrics”

Abstract: “The past few decades have witnessed rapid and unprecedented theoretical progress on the science of causal inference, ranging from the "credibility revolution” with the popularization of quasi-experimental designs in statistics and econometrics, to the development of a complete solution to non-parametric identification with causal graphical models in computer science. Most of this theoretical progress, however, relies on strong, exact assumptions, such as the absence of unobserved common causes (ignorability assumptions), or the absence of certain direct effects (exclusion restrictions). Unfortunately, more often than not these assumptions are very hard to defend in practice. This leads to two undesirable consequences for applied quantitative work: (i) important research questions may be neglected, simply because they do not exactly match the requirements of current methods; or, (ii) researchers may succumb to making the required “identification assumptions” simply to justify the use of available methods, but not because these assumptions are truly believed (or understood).  In this talk, I will discuss new theory, methods, and software for permitting causal inferences under more flexible and realistic settings. In particular, I will focus on a flexible suite of sensitivity analysis tools for OLS (Cinelli and Hazlett, 2020) and instrumental variables (Cinelli and Hazlett, 2021) which can be immediately put to use to improve the robustness and transparency of current applied research. Notably, by building upon the familiar "omitted variable bias" framework, these tools: (i) do not require assumptions on the functional form of the treatment assignment mechanism nor on the distribution of the unobserved variables; (ii) naturally handle multiple unobserved variables, possibly acting non-linearly; (iii) exploit expert knowledge to bound sensitivity parameters; and, (iv) can be easily implemented using standard regression output. Finally, time permitting, I will discuss recent developments of a simple, yet general theory of omitted variable bias in Causal Machine Learning (Chernozhukov et al, 2022).”