Registro completo |
Provedor de dados: |
31
|
País: |
United States
|
Título: |
How Much Can Outlook Forecasts be Improved? An Application to the U.S. Hog Market
|
Autores: |
Colino, Evelyn V.
Irwin, Scott H.
Garcia, Philip
|
Data: |
2008-06-26
|
Ano: |
2008
|
Palavras-chave: |
Forecast
Futures
Models
Prices
Time-series
Vector autoregression
Agricultural Finance
|
Resumo: |
This study investigates the predictability of outlook hog price forecasts released by Iowa State University relative to alternative market and time-series forecasts. The findings suggest that predictive performance of the outlook hog price forecasts can be improved substantially. Under RMSE, VARs estimated with Bayesian procedures that allow for some degree of flexibility and model averaging consistently outperform Iowa outlook estimates at all forecast horizons. Evidence from the encompassing tests, which are highly stringent tests of forecast performance, indicates that many price forecasts do provide incremental information relative to Iowa. Simple combinations of these models and outlook forecasts are able to reduce forecast errors by economically significant levels. The value of the forecast information is highest at the first horizon and then gradually declines.
|
Tipo: |
Conference Paper or Presentation
|
Idioma: |
Inglês
|
Identificador: |
http://purl.umn.edu/37620
|
Relação: |
NCCC-134 Conference on Applied Commodity Price Analysis, Forecasting, and Market Risk Management>2008 Conference, April 21-22, 2008, St. Louis, Missouri
2008 NCCC-134 Conference on Applied Commodity Price Analysis, Forecasting, and Market Risk Management
|
Formato: |
25
|
|