We develop the dual implications of the von Liebig‐Paris nutrient‐response technology and parametric and nonparametric tests for this technology. We discuss the nonparametric approach using two different representations corresponding to optimistic and pessimistic nonparametric approximations of the technology. Application of the nonparametric approach to two experimental data sets and to two data sets generated by Monte Carlo methods confirms the presence of yield plateaus, but provides mixed evidence regarding input nonsubstitution. The optimistic representation exhibits virtually no input substitution, while the pessimistic representation shows considerable substitutability.
No takes yet. Share an insight, caveat, or question.
Chambers et al. (1996) studied this question.