Time-Series Methods for Forecasting and Modeling Uncertainty in the Food Price Outlook

• 查看报告

英文名称：

Time-Series Methods for Forecasting and Modeling Uncertainty in the Food Price Outlook

作者：

Matthew MacLachlan""url":0; Carolyn Chelius""url":0; Gianna Short""url":0;

来源地址：

https://www.ers.usda.gov/api/publications/v1.0?page=6&items_per_page=20&sort_by=releaseDate&sort_order=DESC&series=AP%2CCCR%2CEB%2CEIB%2CERR%2CTB

关键词：

Autoregressive integrated moving average (ARIMA);  Consumer Price Index (CPI);  food prices;  Food Price Outlook (FPO);  forecasting;  model selection;  Producer Price Index (PPI);  time-series econometrics;  uncertainty; 

年份：

2022

出版地：

Washington, DC, USA

总页数：

32

语种：

English

摘要：

The USDA, Economic Research Service’s Food Price Outlook (FPO) provides monthly forecasts of annual food price percent changes up to 18 months in advance. The forecasts add value to the U.S. Bureau of Labor Statistics’ Consumer and Producer Price Indexes (CPI, PPI) by giving farmers, wholesalers, retailers, institutional buyers, consumers, and policymakers a uniform set of predictions about food prices. The more accurate the predictions, the more value FPO contributes. Events such as recent natural disasters, the Great Recession, the Food Crisis of 2011, and the COVID-19 pandemic have highlighted the importance of food price forecasting and the need for improvements to the forecasting methodology to enhance accuracy and treat uncertainty more rigorously. This technical bulletin describes a time-series-based approach for forecasting food prices which provides enhanced precision, removes potential biases from the specification process, and allows for a clearer characterization of uncertainty about future food prices. Four case studies are included to illustrate how these forecasts can be used.