 Assessing the Accuracy of USDA’s Farm Income Forecasts: The Impact of ARMSTodd H. Kuethe and Todd Hubbs No 285874, 2017 Conference, April 24-25, 2017, St. Louis, Missouri from NCR-134/ NCCC-134 Applied Commodity Price Analysis, Forecasting, and Market Risk Management Abstract: The USDA’s forecasts for net farm income are important inputs for businesses, legislators, economists, and other policy-makers. While the USDA has been providing forecasts for net farm income for over 50 years, they have not been rigorously analyzed in regards to bias and efficiency. Here, the USDA’s net farm income forecasts from 1975-2015 are evaluated along a number of dimensions. The results show that the USDA’s initial forecasts for net farm income are downward bias. This bias is corrected as the forecasts evolve through the year. Moreover, upward revisions tend to lead to negative forecast errors or overestimates. Consistent with that finding, there is a tendency for reversals in the forecasts. That is, upward revisions tend to be corrected in subsequent revisions. Despite these inefficiencies, the forecasts provide remarkably good directional guidance by predicting growth or contraction correctly over 80% of the time. There was no evidence that forecast accuracy improved through time or with the availability of ARMS data. Keywords: Marketing (search for similar items in EconPapers) Downloads: (external link) Related works: Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text Persistent link: https://EconPapers.repec.org/RePEc:ags:n13417:285874 Access Statistics for this paper More papers in 2017 Conference, April 24-25, 2017, St. Louis, Missouri from NCR-134/ NCCC-134 Applied Commodity Price Analysis, Forecasting, and Market Risk Management |
Is your work missing from RePEc? Here is how to contribute.
Questions or problems? Check the EconPapers FAQ or send mail to .
EconPapers is hosted by the
School of Business at Örebro University.