ENSAE Paris - École d'ingénieurs pour l'économie, la data science, la finance et l'actuariat

Macroeconometrics and Machine Learning

Enseignant

Objectif

Large and alternative data sets are widely used for macroeconomic forecasting/nowcasting together with machine learning-based tools. This course presents state-of-the-arts methods for macroeconomic forecasting and nowcasting in presence of large data sets. In particular, the methods illustrated during the course are useful to deal with large-dimensional data sets (big data), to deal with data sets with missing observations, and to estimate forecasting models that are potentially complex and nonlinear. Choice of tuning parameters and how this choice affects the accuracy of the forecasting will be discussed.

At the end of the course students will be able to select the most appropriate techniques for macroeconomic forecasting estimation depending on the data available and on the assumptions made about the forecasting model.

Plan

Part I: Factor Models

  • Static and Dynamic Factor Models (FMs)
  • Estimation of the Factors and Parameters
  • Determination of the number of factors
  • Dynamic FMs for Macroeconomic Monitoring and Forecasting

Part II: Machine Learning (ML) methods in Macroeconometrics

  • Introduction: data-poor versus data-rich environments, features of ML.
  • Penalized linear models, dense versus sparse models.
  • Tree-based methods and Random forests.
  • Neural Networks.
  • Explainable machine learning

Part III: Vector Autoregression

  •   General properties of stationary vector processes. Innovations process.
  • Stationary VAR process and invertibility. Innovations process. Forecasting with a VAR.
  • Estimation of a VAR model. Tests.
  • Information criteria.
  • Propagation of shocks : impulse response functions. Introduction to structural VARs.

Références

Part I and II of the course will not follow a textbook. References to several academic articles will be provided during the course.

 

For Part III:

BROCKWELL P.J. et DAVIS R.A. (1990). Time Series. Theory and Methods. Springer-Verlag.

GOURIEROUX C. et MONFORT A. (1995). Séries temporelles et modèles dynamiques, 2ème ed., Economica.

HAMILTON J.D. (1994). Time Series Analysis, Princeton Univ. Press.

JOHANSEN S. (1995). Likelihood-based inference in cointegrated Vector Auto-Regressive models, Oxford University Press.

JUSELIUS K. (2006). The cointegrated VAR model, Oxford Univ Press.

KILIAN L., LÜTKEPOHL H. (2017), Structural Vector Autoregressive Analysis, Cambridge University Press

LÜTKEPOHL H. (2005). New Introduction to Multiple Time Series Analysis, Springer Verlag.