Hierarchical Bayesian Estimation of a Stochastic Plateau Response Function: Determining Optimal Levels of Nitrogen Fertilization

Ouédraogo, Frédéric; Brorsen, B. Wade

doi:10.1111/cjag.12139

Cited by 17 publications

(15 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For conventional low-dimensional yield functions, it is common to use simple concave functions such as quadratic [46][47][48], negative exponential [49,50], and piecewise linear [51,52]. These simple functions may serve well for answering isolated questions about the optimality of a single management variable under homogeneous conditions.…”

Section: Farmer's Problemmentioning

confidence: 99%

Machine learning for optimizing complex site-specific management

Saikai

Patel

Mitchell

2020

Computers and Electronics in Agriculture

View full text Add to dashboard Cite

Despite the promise of precision agriculture for increasing the productivity by implementing site-specific management, farmers remain skeptical and its utilization rate is lower than expected. A major cause is a lack of concrete approaches to higher profitability. When involving many variables in both controlled management and monitored environment, optimal site-specific management for such high-dimensional cropping systems is considerably more complex than the traditional low-dimensional cases widely studied in the existing literature, calling for a paradigm shift in optimization of site-specific management. We propose an algorithmic approach that enables farmers to efficiently learn their own site-specific management through on-farm experiments. We test its performance in two simulated scenarios-one of medium complexity with 150 management variables and one of high complexity with 864 management variables. Results show that, relative to uniform management, site-specific management learned from 5-year experiments generates $43/ha higher profits with 25 kg/ha less nitrogen fertilizer in the first scenario and $40/ha higher profits with 55 kg/ha less nitrogen fertilizer in the second scenario. Thus, complex site-specific management can be learned efficiently and be more profitable and environmentally sustainable than uniform management.

show abstract

Section: Farmer's Problemmentioning

confidence: 99%

Machine learning for optimizing complex site-specific management

Saikai

Patel

Mitchell

2020

Computers and Electronics in Agriculture

View full text Add to dashboard Cite

show abstract

“…Both model specifications can be estimated using frequentist techniques for the estimation of non-linear mixed-effects models. Ouedraogo and Brorsen (2017) detail an alternative Bayesian approach.…”

Section: Conceptual and Econometric Frameworkmentioning

confidence: 99%

Estimating Economically Optimal Levels of Nitrogen Fertilizer in No-Tillage Continuous Corn

Villacis

Ramsey

Delgado

et al. 2020

J. Agric. Appl. Econ.

View full text Add to dashboard Cite

Stochastic plateau production functions provide improved fertilizer recommendations based on multi-year agronomic experiments where weather and other stochastic variables change over time. This research assesses the profitability of no-tillage corn production in northeastern Colorado and determines economically optimal nitrogen fertilizer rates. It also proposes an alternative parameterization of the linear response stochastic plateau model which provides a robustness check against traditional parameterizations. Results show the current use of nitrogen fertilizer in the area exceeds estimated economically optimal levels. This suggests that a reduction in nitrogen use could increase expected profits and simultaneously reduce environmental costs.

show abstract

“…This implies that a vector of parameters of interest is actually: = ( , , , , )′. Following (Ouedraogo and Brorsen 2018), the Bayesian approach to providing the solution to equation (4) involves a maximization of equation (13) 13)…”

Section: {Figure 31}mentioning

confidence: 99%

Economics of the Greenseeder Hand Planter, Discrete Choice Modeling, and On-Farm Field Experimentation

Ng’ombe¹

2019

Preprint

View full text Add to dashboard Cite

Corn yields in developing countries are lower than in developed countries partly due to planting methods that involve hand dropping of multiple seeds per hill. Researchers from Oklahoma State University (OSU) recently developed the Greenseeder Hand Planter (GHP) to replace such methods. The first essay determines economic breakeven levels of seed and labor savings, increases in corn yield, and reduced loss of N through reduced ammonia volatilization. Results suggest a $50 GHP used to plant 3 hectares per year would need to increase corn yields on average by about 1.12%, use 12.19% fewer seeds, or reduce labor man-days by 38.66% to equal expected net returns from traditional methods.In the second essay, I conduct Monte Carlo experiments to measure bias in the conditional logit (CL) and independent availability logit (IAL) when there is no choice set formation and when choice sets are stochastically formed. I also compare the performance of the two models using empirical data on paddlefish angler preferences collected in Oklahoma. Both the CL and IAL work well when their own assumptions hold, but not under the alternative’s assumptions. However, the IAL produces unbiased and less efficient parameter estimates when individuals actually choose from the full set of alternatives. Empirical results suggest the IAL is able to predict the attribute-cutoff.To avoid limitations from small-scale agronomic trials, there has been a movement toward large-scale, on-farm field trials but questions remain as how best to conduct them and when it is most profitable to quit them. The third essay addresses these questions by using a fully Bayesian decision-theoretic approach. Data are from Monte Carlo simulations assuming a corn-input stochastic plateau production function. Results suggest the best way to conduct such experiments is to allocate to each of the 10% of the plots, 0 lb. of N, half of N*, and 150% of N* under a 30-plot experimental design. Results further indicate that it optimal to quit such trials in year 2. Sensitivity analysis confirms the optimal quit period but suggests such experiments are most profitable by allocating unalike N levels to all of the 30% of experimental plots.

show abstract

Hierarchical Bayesian Estimation of a Stochastic Plateau Response Function: Determining Optimal Levels of Nitrogen Fertilization

Cited by 17 publications

References 34 publications

Machine learning for optimizing complex site-specific management

Machine learning for optimizing complex site-specific management

Estimating Economically Optimal Levels of Nitrogen Fertilizer in No-Tillage Continuous Corn

Economics of the Greenseeder Hand Planter, Discrete Choice Modeling, and On-Farm Field Experimentation

Contact Info

Product

Resources

About