What is it about?
This paper is a survey of quantile econometrics. It explains how economists can study not just the average effect of a variable, but what happens at different points of the distribution, especially in the lower and upper tails, where risk and extreme outcomes matter most. The article reviews the main methods in the literature, including cross-sectional quantile regression, time-series quantile models, quantile VAR, panel quantile models, and factor-augmented quantile models such as QFAVAR, and shows how these approaches help capture heterogeneity, asymmetry, nonlinearities, and tail risk that standard mean-based models often miss. Overall, the paper argues that quantile methods give policymakers and researchers better tools for understanding economic and financial risks across the full distribution of outcomes, not only at the center.
Featured Image
Photo by Bozhin Karaivanov on Unsplash
Why is it important?
It is important because the paper argues that standard econometric models focus only on the average effect, which can hide what is happening in the best, worst, and most vulnerable parts of the distribution. Quantile econometrics is valuable because it shows how relationships change across the bottom, middle, and top outcomes, making it especially useful for studying tail risk, heterogeneity, asymmetry, and nonlinear effects. This matters in real economic problems such as credit losses, unemployment, inflation spikes, productivity slowdowns, and financial stress, where policymakers and regulators need tools that are robust to outliers and useful for decisions under extreme conditions. In that sense, the paper sees quantile methods as important because they help explain not just what happens on average, but who is most affected, when risks become severe, and how policy can be designed across the full range of possible outcomes.
Perspectives
From a future-perspectives point of view, the paper is important because it shows that quantile econometrics is not a finished field but a growing framework with major room for development in how economists measure extreme risks and distributional effects. The authors argue that the next advances should extend quantile methods to richer settings such as distributional Difference-in-Differences, quantile factor models, and especially a quantile Global VAR (Q-GVAR), which could trace tail risks and spillovers across countries, sectors, firms, and markets. This matters because modern economic risks are increasingly systemic, nonlinear, and concentrated in the tails, so future models must go beyond averages to capture cross-country linkages, time-varying tail dependence, and high-dimensional data in a coherent way. In that sense, the paper’s perspective is that quantile econometrics will become even more valuable for macro-financial surveillance, policy design, and global risk monitoring as researchers solve these methodological and computational challenges.
Dr. Dimitrios Konstantios
Alba Graduate Business School, The American College of Greece
Read the Original
This page is a summary of: Econometrics at the Extreme: From Quantile Regression to QFAVAR
1, Journal of Economic Surveys, January 2026, Wiley,
DOI: 10.1111/joes.70063.
You can read the full text:
Resources
Contributors
The following have contributed to this page
