- Consulting Editors
- Vars, Cointegration, and Common Cycle Restrictions
- Dynamic Factor Models
- Forecasting With Nonlinear Time Series Models
- Forecasting With DSGE Models
- Forecasting Economic Time Series Using Unobserved Components Time Series Models
- Improving the Role of Judgment in Economic Forecasting
- Forecasting with Mixed-Frequency Data
- Forecasting with Real-Time Data Vintages
- Forecasting from misspecified Models in the Presence of Unanticipated Location Shifts
- Forecasting Breaks and Forecasting During Breaks
- Forecast Combinations
- Multiple Forecast Model Evaluation
- Testing for Unconditional Predictive Ability
- Testing Conditional Predictive Ability
- Interpreting and Combining Heterogeneous Survey Forecasts
- Analyzing Three-Dimensional Panel Data of Forecasts
- Forecasting Financial Time Series
- Forecasting Volatility Using High-Frequency Data
- Economic Value of Weather and Climate Forecasts
- Long-Horizon Growth Forecasting and Demography
- Forecasting the Energy Markets
- Models for Health Care
- Election Forecasting
- Marketing and Sales
Abstract and Keywords
This article discusses out-of-sample evaluation based on assessing the conditional predictive ability of forecasting models, which stands in contrast to the notion of unconditional predictive ability testing. Here the word “conditional” refers to any approach in the literature that has gone beyond assessing the forecasting performance of models on average. The common thread which unites the papers discussed in this article is the argument that focusing solely on the average, or global, performance of a model may result in a loss of information and possibly lead to incorrect forecast selection decisions. Most of the discussion focuses on the problem of comparing the performance of two competing forecasts, but some extensions to different testing environments are also pointed out.
Raffaella Giacomini is a reader (associate professor) of economics at University College, London. Her research interests are time series econometrics and forecasting, with an emphasis on forecast evaluation, model selection, and estimation of dynamic general equilibrium models with applications to macroeconomic data.
Access to the complete content on Oxford Handbooks Online requires a subscription or purchase. Public users are able to search the site and view the abstracts and keywords for each book and chapter without a subscription.
If you have purchased a print title that contains an access token, please see the token for information about how to register your code.