Time Series Econometrics Overview

Principles

Definition

Time series econometrics is a specialized area of econometrics concerned with the development and application of statistical methods for analyzing data indexed by time. This field examines temporal dependencies, persistence, and causality in economic and financial variables to model, forecast, and test hypotheses about dynamic processes.

Ontological type

Core Methods

Key Assumptions

Economic Applications

Key developments

Key Figures

Nonstationarity and Cointegration era

Robert F. Engle [1] is a leading economist associated with Harvard University [3] and the Massachusetts Institute of Technology [4] during this era. Robert F. Engle [1] championed the co-integration and error-correction framework as described in Co-Integration and Error Correction: Representation, Estimation, and Testing [7], and his work on Seasonal integration and cointegration [8] clarified the empirical machinery for nonstationary time series in this era. Søren Johansen [2] was affiliated with Imperial College London [5] and the University of Copenhagen [6] during this era. Søren Johansen [2] authored Maximum Likelihood Estimation and Inference on Cointegration — with Applications to the Demand for Money [9] and Estimation and Hypothesis Testing of Cointegration Vectors in Gaussian Vector Autoregressive Models [10], which formalized ML-based cointegration inference and vector error-correction representations that underpinned analysis in near-unit-root systems.

Structural Change and Time Variation era

Jushan Bai [1] conducted influential work during this era at the Massachusetts Institute of Technology [3] and Columbia University [4]. His key contribution, through the 1998 paper Estimating and Testing Linear Models with Multiple Structural Changes [7], is to provide procedures for estimating and testing linear models with multiple structural breaks, enabling robust inference about regime shifts and evolving dynamics in time series. Pierre Perrón [2] has been affiliated with Boston University [5] and Boston College [6] during this era. Perrón [2] contributed to the same paper, Estimating and Testing Linear Models with Multiple Structural Changes [7], advancing break detection and inference for time-varying and regime-switching dynamics that were central to the era.

Real-time Causal Forecasting era

Nicholas Bloom [1] is associated with Harvard University [3] and Stanford University [4] during this era. He contributed to real-time policy uncertainty analysis by developing the Measuring Economic Policy Uncertainty* [6], a result that shows how policy uncertainty shapes macro outcomes and forecast credibility in this era of time-varying inference. Steven J. Davis [2] is affiliated with Stanford University [4] and the University of Pennsylvania [5] during this era. Davis [2] co-authored the Measuring Economic Policy Uncertainty* [6], helping to establish a real-time uncertainty measure that informs horizon-specific inference and policy evaluation amid revised and high-frequency data.