Location: Soil Management and Sugarbeet ResearchTitle: Estimating economically optimal levels of nitrogen fertilizer in no-tillage continuous corn
|VILLACIS, ALEXIS - VIRGINIA TECH|
|RAMSEY, FORD - VIRGINIA TECH|
|ALWANG, JEFFREY - VIRGINIA TECH|
Submitted to: Journal of Agricultural and Applied Economics
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/25/2020
Publication Date: 10/14/2020
Citation: Villacis, A.H., Ramsey, F.A., Delgado, J.A., Alwang, J.R. 2020. Estimating economically optimal levels of nitrogen fertilizer in no-tillage continuous corn. Journal of Agricultural and Applied Economics. 52(4):613-623. https://doi.org/10.1017/aae.2020.23.
Interpretive Summary: We estimate corn yield response functions to nitrogen using a switching regression plateau model and the linear stochastic plateau models of Tembo et al. (2008) and Tumusiime et al. (2011). The empirical application uses data from a long-term study of corn yields conducted at the Agricultural Research Development and Education Center (ARDEC) located in Fort Collins, Colorado. Our results show that under equivalent assumptions on the random parameters and correlation of the random effects, the switching regression model(s) performs better than the specification of Tembo et al. (2008) and as good as the specification of Tumusiime et al. (2011). We suggest the switching regression specification as a useful approach for applied analyses of crop response.
Technical Abstract: We propose a switching regression (SR) approach for estimating crop yield response models with stochastic plateaus. The SR method treats the minimum input level necessary to reach the yield plateau as a model parameter. We find that the SR specification compares favorably to other stochastic plateau models suggested in the literature in terms of statistical fit criteria and economic significance. An empirical application to data from a no-tillage continuous corn experiment in Colorado demonstrates the process of deriving nitrogen recommendations and profit estimates and illustrates the proposed methodology.