FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2038581084> ?p ?o ?g. }

• W2038581084 endingPage "3803" @default.
• W2038581084 startingPage "3789" @default.
• W2038581084 abstract "Abstract Conventional procedures for calculating confidence limits of forecasts generated by statistical models provide little guidance for forecasts based on a combination or a consensus process rather than formal models, as is the case with US Department of Agriculture (USDA) forecasts. This study applied and compared several procedures for calculating empirical confidence intervals for USDA forecasts of corn, soybean and wheat prices over the 1980/81 through 2006/07 marketing years. Alternative procedures were compared based on out-of-sample performance over 1995/96 through 2006/07. The results of this study demonstrate that kernel density, quantile distribution and best fitting parametric distribution (logistic) methods provided confidence intervals calibrated at the 80% level prior to harvest and 90% level after harvest. The kernel density-based method appears most accurate both before and after harvest with the final value falling inside the forecast interval 77% of the time before harvest and 92% after harvest, followed by quantile regression (73% and 91% before and after harvest, respectively) logistic distribution (73% and 90% before and after harvest, respectively) and histogram (66% and 84% before and after harvest, respectively). Overall, this study demonstrates that empirical approaches may be used to construct more accurate confidence intervals for USDA corn, soybean and wheat price forecasts. Acknowledgements The funding support of the USDA under cooperative Agreement 43-3AEK-5-80076 is gratefully acknowledged. Any opinions, findings, conclusions or recommendations expressed in this publication are those of the authors and do not necessarily reflect the view of the USDA. Notes 1 It is worth noting that most theoretical variance expressions are based on the same assumption. 2 Isengildina et al. (Citation2004) provide survey evidence that USDA price intervals are symmetric. That is, a midpoint is forecast and then an equal interval is added to each side of the midpoint. Therefore, the average forecast price is computed in this study by taking an average of the midpoint of forecast prices for each month. 3 The last several months (17 and 18 for corn and soybeans and 14 and 15 for wheat) were not included in the analysis because errors were often zero, so the distributions were impossible to estimate. 4 It is important to note that the forecasting cycle for each of the commodities included in this analysis exceeds one calendar year. This implies that the final estimates (and thus forecast errors) for the previous year are not known until the 6th month into the next forecasting cycle for corn and soybeans and the 4th month for wheat. Therefore, the estimation sub-sample is one observation smaller in these early months of the forecasting cycle, i.e. while most intervals for 2006/07 are constructed using 26 prior observations, only 25 observations are available for the first 5 months of the corn and soybean forecasting cycle and the first 3 months of the wheat forecasting cycle. 5 This procedure was used by Isengildina et al. (Citation2004) to construct 95% confidence level forecast intervals and implied confidence level forecast intervals as described in footnote 9 of their paper. 6 The bin width of 0.05% was selected for this study, which was the finest bin based on the data, the coarser bin could have resulted in wider intervals. 7 Other distributions, such as gamma and Weibull were not included in this analysis because they are not defined for negative values. 8 To calculate the chi-squared statistic, one must break up the x-axis domain into ‘bins’ or classes of data. However, there are no clear guidelines for selecting the number and location of the bins and different conclusions can be reached from the same data depending on how bins are specified. 9 Computations were also made where the best-fitting distribution was allowed to change from year-to-year for corn. These alternative computations yielded quite similar results to those limited to the logistic distribution and are available from the authors upon request. 10 It is important to keep in mind that these interval widths are conditional on the selected confidence levels (80–90%). If the selected confidence level was equal to the actual hit rates, the widths of the empirical confidence intervals would be approximately equal to that of published intervals. 11 There is some evidence that mean forecast error of corn prices prior to harvest also shifted to the right, but the shift was not statistically significant. 12 Christoffersen (Citation1998) also proposed additional tests that examine interval forecast independence and forecast coverage conditional on independence. These tests, however, should only be applied across marketing years. For example, the May forecast for 2005/06 should be independent from the May forecast for 2006/07. Independence, however, is not possible between April and May forecasts for 2006/07, as these are forecasts of the same event. Therefore, our approach of combining pre- and post-harvest forecasts for accuracy analysis would yield meaningless results due to violation of independence across forecasts of the same events. Thus, due to a small number of observations, these tests cannot be applied reliably to the prediction part of the sample (1995/96–2006/07)." @default.
• W2038581084 created "2016-06-24" @default.
• W2038581084 creator A5009247199 @default.
• W2038581084 creator A5052053188 @default.
• W2038581084 creator A5069018102 @default.
• W2038581084 creator A5079409910 @default.
• W2038581084 date "2011-10-01" @default.
• W2038581084 modified "2023-09-26" @default.
• W2038581084 title "Empirical confidence intervals for USDA commodity price forecasts" @default.
• W2038581084 cites W1590041868 @default.
• W2038581084 cites W1977635687 @default.
• W2038581084 cites W1977970167 @default.
• W2038581084 cites W2019698004 @default.
• W2038581084 cites W2043250042 @default.
• W2038581084 cites W2046925783 @default.
• W2038581084 cites W2063572227 @default.
• W2038581084 cites W2117414963 @default.
• W2038581084 cites W2140349953 @default.
• W2038581084 cites W2154355633 @default.
• W2038581084 cites W2169658553 @default.
• W2038581084 cites W2488358200 @default.
• W2038581084 cites W3125396275 @default.
• W2038581084 cites W4233014035 @default.
• W2038581084 cites W4235016709 @default.
• W2038581084 cites W4241333530 @default.
• W2038581084 cites W4248204858 @default.
• W2038581084 cites W4253661766 @default.
• W2038581084 doi "https://doi.org/10.1080/00036841003724429" @default.
• W2038581084 hasPublicationYear "2011" @default.
• W2038581084 type Work @default.
• W2038581084 sameAs 2038581084 @default.
• W2038581084 citedByCount "14" @default.
• W2038581084 countsByYear W20385810842013 @default.
• W2038581084 countsByYear W20385810842014 @default.
• W2038581084 countsByYear W20385810842016 @default.
• W2038581084 countsByYear W20385810842017 @default.
• W2038581084 countsByYear W20385810842018 @default.
• W2038581084 countsByYear W20385810842019 @default.
• W2038581084 countsByYear W20385810842020 @default.
• W2038581084 countsByYear W20385810842021 @default.
• W2038581084 countsByYear W20385810842022 @default.
• W2038581084 countsByYear W20385810842023 @default.
• W2038581084 crossrefType "journal-article" @default.
• W2038581084 hasAuthorship W2038581084A5009247199 @default.
• W2038581084 hasAuthorship W2038581084A5052053188 @default.
• W2038581084 hasAuthorship W2038581084A5069018102 @default.
• W2038581084 hasAuthorship W2038581084A5079409910 @default.
• W2038581084 hasConcept C105795698 @default.
• W2038581084 hasConcept C118671147 @default.
• W2038581084 hasConcept C149782125 @default.
• W2038581084 hasConcept C151956035 @default.
• W2038581084 hasConcept C162324750 @default.
• W2038581084 hasConcept C185429906 @default.
• W2038581084 hasConcept C33923547 @default.
• W2038581084 hasConcept C44249647 @default.
• W2038581084 hasConcept C63817138 @default.
• W2038581084 hasConcept C71134354 @default.
• W2038581084 hasConcept C98385598 @default.
• W2038581084 hasConceptScore W2038581084C105795698 @default.
• W2038581084 hasConceptScore W2038581084C118671147 @default.
• W2038581084 hasConceptScore W2038581084C149782125 @default.
• W2038581084 hasConceptScore W2038581084C151956035 @default.
• W2038581084 hasConceptScore W2038581084C162324750 @default.
• W2038581084 hasConceptScore W2038581084C185429906 @default.
• W2038581084 hasConceptScore W2038581084C33923547 @default.
• W2038581084 hasConceptScore W2038581084C44249647 @default.
• W2038581084 hasConceptScore W2038581084C63817138 @default.
• W2038581084 hasConceptScore W2038581084C71134354 @default.
• W2038581084 hasConceptScore W2038581084C98385598 @default.
• W2038581084 hasIssue "26" @default.
• W2038581084 hasLocation W20385810841 @default.
• W2038581084 hasOpenAccess W2038581084 @default.
• W2038581084 hasPrimaryLocation W20385810841 @default.
• W2038581084 hasRelatedWork W2009605178 @default.
• W2038581084 hasRelatedWork W2052558588 @default.
• W2038581084 hasRelatedWork W2079454017 @default.
• W2038581084 hasRelatedWork W2110394835 @default.
• W2038581084 hasRelatedWork W2171769143 @default.
• W2038581084 hasRelatedWork W2224749288 @default.
• W2038581084 hasRelatedWork W2606485752 @default.
• W2038581084 hasRelatedWork W3128481383 @default.
• W2038581084 hasRelatedWork W4237949580 @default.
• W2038581084 hasRelatedWork W4324319436 @default.
• W2038581084 hasVolume "43" @default.
• W2038581084 isParatext "false" @default.
• W2038581084 isRetracted "false" @default.
• W2038581084 magId "2038581084" @default.
• W2038581084 workType "article" @default.