Clustering and Forecasting Inflation Expectations Using the World Economic Survey: The Case of the 2014 Oil Price Shock on Inflation Targeting Countries


  • Borradores de economía


  • This paper examines inflation expectations of the World Economic Survey for ten inflation targeting countries. First, by a Self Organizing Maps methodology, we cluster the trajectory of agent’s inflation expectations using the beginning of the oil price shock occurred in June of 2014 as a benchmark in order to discriminate between those countries that anticipated the shock smoothly and those with brisk changes in expectations. Then, the expectations are modeled by artificial neural networks forecasting models. Second, for each country we investigate the information content of the quantitative survey forecast by comparing it to the average annual inflation based on national consumer price indices. The results indicate the presence of heterogeneity among countries to anticipate inflation under the oil shock and, also different patterns of accuracy to predict average annual inflation were found depending on the observed inflation trend.

fecha de publicación

  • 2017-05

Líneas de investigación

  • Inflation Expectations
  • Machine Learning
  • Nonlinear Autoregressive Neural Network
  • Self Organizing Maps


  • 993