Modelling Structural Breaks


A key aspect of the financial crisis was the lack of fore-knowledge, so analyses of forecasting in the face of breaks, during breaks and of breaks themselves remain a central focus. In a world where breaks are unanticipated and occur intermittently, it is easy to understand why economic forecasting has a poor track record of missing major changes, and being systematically wrong for prolonged periods. We are developing improved forecasting methods; approaches to try and forecast breaks, or failing that, forecasting during breaks; and building robust forecasting devices. Importantly, if agents were to use such devices, a new basis for both economics and empirical models must be forged. New data sources, such as Google Trends, provide disaggregated high-frequency information on a vast scale, both allowing the possibility of improved ‘nowcasts’ for flash estimates of GNP, inflation etc., and providing incipient information on agents’ reactions to changes to help predict breaks and during breaks.


Project Leader / Primary Investigator

Prof. Sir David Hendry

Related Library Items

Forecasting Breaks and During Breaks
On not evaluating economic models by forecast outcomes
Forecasting from misspecified Models in the Presence of Unanticipated Location Shifts
Combining Disaggregate Forecasts or Combining Disaggregate Information to Forecast an Aggregate
Open-Model Forecast-Error Taxonomies
Economic fundamentals and Eurozone sovereign spreads: will the good news continue?
Introduction to Special Section on Exchange Rate Pass-through in Developing and Emerging Markets
The aggregate mortgage possessions outlook
Forecasting by factors, by variables, by both or neither?
Detecting Location Shifts During Model Selection by Step-indicator Saturation
Detecting Big Structural Breaks in Large Factor Models
US Inflation Expectations and Heterogeneous Loss Functions, 1968–2010
Are Professional Macroeconomic Forecasters Able To Do Better Than Forecasting Trends?
Forecast Uncertainty—Ex Ante and Ex Post: U.S. Inflation and Output Growth
Probability Distributions or Point Predictions? Survey Forecasts of US Output Growth and Inflation
Forecasting with Vector Autoregressive Models of Data Vintages: US Output Growth and Inflation
Real-time Forecasting of Inflation and Output Growth with Autoregressive Models in the Presence of Data Revisions
Improving Real-Time Estimates of Output and Inflation Gaps With Multiple-Vintage Models
How Good are US government Forecasts of the Federal debt?
Forecasting with Bayesian Multivariate Vintage-Based VARs'
Assessing the Evidence of Macro-Forecaster Herding: Forecasts of Inflation and Output Growth
Anticipating early data revisions to US GDP and the effects of releases on equity markets
2014 Forecast Scenarios for UK Mortgage Arrears and Possessions
New forecast scenarios for UK mortgage arrears and possessions
Modelling and Forecasting with County Court Data: Regional Mortgage Possession Claims and Orders in England and Wales
Generalised empirical likelihood-based kernel density estimation
Do US macroeconomic forecasters exaggerate their differences?
Real-Time Factor Model Forecasting and the Effects of Instability
New methods for forecasting inflation, applied to the USA
Exchange Rate Pass-through to Import Prices and Monetary Policy in South Africa
The real-wage productivity nexus
Evaluating multi-step system forecasts with relatively few forecast-error observations
Quantifying the Uncertainty around Break Dates in Models using Indicator Saturation

All Projects