Much research shows that the ratings that critics, judges, and consumers assign to wines are heteroscedastic. A rating observed is one draw from a latent distribution that is wine- and judge-specific. Estimating the shape of a rating’s distribution by minimizing a sum of cross entropies has been proposed and tested. This article proposes a method of improving the accuracy of that estimate by using information about the context of a wine competition or cross-section ratings data. Tests using the distributions implied by 90 blind triplicate ratings show that the sum of squared errors for the solution using context or cross-section information is 50% more accurate than not using such information and over 99% more accurate than ignoring the uncertainty about a rating.

Analyses and aggregations of the ratings that wine critics and judges assign to wines are made difficult by stochastic error and biases that remain even when wines are assessed blind to price, label, capsule, and closure. Stochastic error is due to the partially random nature of ratings. Cognitive and omitted-variable biases are due to anchoring, expectation, serial position, commercial, and other factors. Differences in decanting, filtering, aeration, and temperature can also affect ratings.

Much research shows that the ratings that judges assign to the same wine are uncertain. And while the ratings may be independent, research also shows that they are not identically distributed. Thus, an acute difficulty in ratings-related research and in calculating consensus among judges is that each rating is one observation drawn from a latent distribution that is wine- and judge-specific. What can be deduced about the shape of a latent distribution from one observation? A simple maximum entropy estimator is proposed to describe the distribution of a rating observed. The estimator can express the implications of zero, one, a few blind replicates, and many observations. Several tests of the estimator show that results are consistent with the results of experiments with blind replicates and that results are more accurate than results based on observed ratings alone.

There can be a serious tension between the commitment to cost-benefit analysis and a realistic appreciation of the limits of official knowledge. Without significant efforts to reduce those limits, that analysis might be inadequately informed. Whenever regulators face significant informational deficits, or what is sometimes called “the knowledge problem,” it is important to explore tools that take advantage of what the private sector knows; market-friendly tools, such as economic incentives, have important advantages on that count. An advanced regulatory system should also try to reduce the knowledge problem through three routes: (i) creative use of notice-and-comment rulemaking; (ii) retrospective analysis of regulations and their costs and benefits; and (iii) advance testing, as a way of informing ex ante analysis. For the future, the most promising approach is (iii).

In a previously published article, I reported some tests of prospect theory's reflection effect over outcomes defined by money and life years gained from treatment. Those results suggested qualified support for the reflection effect over money outcomes and strong support over longevity outcomes. This article reruns those tests while accounting for the intensity of individual risk attitudes, and, overall, show consistency with the reflection effect. However, I argue that these results do not necessarily offer support for the explanatory power of prospect theory. Rather, the results may be driven by evolved responses to circumstances that provoke perceptions of scarcity and abundance. Therefore, from an ecological perspective, behavioral patterns such as those that are consistent with the reflection effect, which, by extension, tend to be considered as erroneous or biased by most behavioral economists because they conflict with the postulates of rational choice theory, may not be unreasonable. Recognizing as such is important when considering how behavioral insights ought to inform public policy design and implementation.

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.

Deterministic and stochastic yield response plateau functions were estimatedto determine the expected profit-maximizing nitrogen rates, yields, and netreturns for corn grown after corn, cotton, and soybeans. The stochasticresponse functions were more appropriate than their deterministiccounterparts, and the linear response stochastic plateau described the datathe best. The profit-maximizing nitrogen rates were similar for corn aftercorn, cotton, and soybeans, but relative to corn after corn, the expectedcorn yield plateaus increased by 12% and 16% after cotton and soybeans,respectively. Expected net returns increased for corn after cotton andsoybeans relative to corn after corn.

The imputed pig death loss contained in the reported monthly U.S. Department of Agriculture (USDA) pig crop data over the December 1995-June 2006 period ranged from —4.93% to 12.75%. Clearly, there are substantial measurement errors in the USDA monthly pig crop data. In this paper, we present alternative monthly U.S. pig crop data using the biological production process, which is compatible with prior knowledge of the U.S. hog industry. Alternative pig crop data are applied to a slaughter hog model and tested comparatively to USDA pig crop. Test results reject the validity of USDA pig crop data in favor of the alternative data.

A Cox nonnested test is conducted using a fast double bootstrap (FDB) methodto select among three competing functional forms (linear response plateau,quadratic, and Mitscherlich-Baule) to model forage yield response tonitrogen applied with swine effluent. The quadratic is rejected in favor ofone of the other functional forms in all cases. The FDB pvalues differed slightly from the single bootstrap pvalues. Buffalograss was slightly more profitable than bermudagrass and hasthe ability to use almost as much nitrogen as bermudagrass.

The parameters of yield response functions can vary by year. Past studiesusually assume yield functions are nonstochastic or “limited” stochastic. Inthis study, we estimate rye–ryegrass yield functions in which all parametersare random. The three functional forms considered are the linear responseplateau, the quadratic, and the Spillman-Mitscherlich. Nonstochastic yieldmodels are rejected in favor of stochastic parameter models. Quadraticfunctional forms fit the data poorly. Optimal nitrogen applicationrecommendations are calculated for the linear response plateau andSpillman-Mitscherlich. The stochastic models lead to smaller recommendedlevels of nitrogen, but the economic benefits of using fully stochastic cropyield functions are small because expected profit functions are relativelyflat for the stochastic yield functions. Stochastic crop yield functionsprovide a way of incorporating production, uncertainty into inputdecisions.