Reflections on the LSE Tradition in Econometrics: a Student's Perspective

Abstract

Since the mid 1960s the LSE tradition, led initially by Denis Sargan and later by David Hendry, has contributed several innovative techniques and modeling strategies to applied econometrics. A key feature of the LSE tradition has been its striving to strike a balance between the theory-oriented perspective of textbook econometrics and the ARIMA data-oriented perspective of time series analysis. The primary aim of this article is to provide a student’s perspective on this tradition. It is argued that its key contributions and its main call to take the data more seriously can be formally justified on sound philosophical grounds and provide a coherent framework for empirical modeling in economics. Its full impact on applied econometrics will take time to unfold, but the pervasiveness of its main message is clear: statistical models that account for the regularities in data can enhance the reliability of inference and policy analysis, and guide the search for better economic theories by demarcating ‘what there is to explain’.

Keywords

• LSE econometrics
• textbook econometrics
• ARIMA
• model validation
• model selection
• error correction model