计量经济学：2025–2026 年度综述

An annual survey of the econ.EM arXiv category, covering research published from September 2025 – August 2026. Built on a close-reading set of 100 papers and organised into 32 chapters, it maps the year by the field's research sub-directions.

Book framework (share of the year's output):

• Structural Policy Evaluation and Heterogeneous Treatment Effects (88.8%) — This cluster focuses on estimating structural parameters, policy impacts, and heterogeneous treatment effects using large-scale data, Bayesian methods, and network models.
• Causal Inference and Robust Treatment Effect Estimation (6.8%) — This direction addresses identification and estimation challenges in causal inference, including boundary discontinuity designs, endogeneity testing, and debiased estimators for average treatment effects.
• Debiased Machine Learning and Generalized Moment Estimation (1.2%) — This area develops theoretical frameworks and software for debiased machine learning, generalized Riesz regression, and double machine learning to handle high-dimensional covariates and complex dependencies.
• Instrumental Variable Regression and Weak Instrument Inference (1.2%) — This subfield tackles inference in instrumental variable models, focusing on weak instruments, invalid instruments, nonparametric settings, and robust testing procedures.
• Financial Time Series Forecasting and Systemic Risk Analysis (0.8%) — This cluster applies econometric and machine learning techniques to forecast financial prices, assess systemic risk, and evaluate predictive accuracy in electricity and cryptocurrency markets.
• Dynamic Panel Data and Factor Models for Time Series (0.8%) — This direction deals with modeling dynamic panel data, latent factors, and common trends in large-scale time series, including methods for handling contamination and dyadic dependencies.
• Nonparametric Specification Tests and Conditional Inference (0.2%) — This niche area focuses on developing finite-sample specification tests, conditional predictive ability tests, and frameworks for drawing qualitative conclusions from quantitative estimates.

Each sub-direction is covered across three chapters — how it framed its questions, the strength of its key findings, and its open debates and frontiers — so the reader gets a traceable picture of what this part of the field actually produced over the year.