The Strategy Aggregation Effect in Judgment

Predicting a criterion that is probabilistically related to pieces of information, or cues, is a paradigmatic judgment task that has been investigated both, in research trying to identify the individual judgment and decision making strategies people use, and in the wisdom-of-crowds literature where the focus is on how aggregation can improve accuracy. I combine these two lines of research to investigate how the performance of individual and aggregated linear strategies are affected by different environmental and group aggregation factors and how performance differences between individual and aggregated linear strategies can be understood in a unified framework. I show that constrained linear strategies are more accurate for individual judgments, but when these judgments are averaged, an unconstrained linear strategy is more accurate. This strategy aggregation effect can be understood by analyzing a decomposition of the mean squared error into bias, variance, and covariance. Because of their lower bias but higher variance, unconstrained linear strategies perform worse for individual judgments, but better for averaged judgments where aggregation minimizes variance. In simulations with artificial and real environments, I further show that this aggregation effect does not occur if there are correlations between individual judgments. Here, constrained linear strategies always outperform an unconstrained linear strategy, because the larger covariance component of the unconstrained linear strategy outweighs its lower bias. I end with real-world implications of the results for cognitive strategies and decision environments in group and organizational settings.

Using portfolio theory in spatial targeting of forest carbon payments: an effective strategy to address spatiotemporal variation in land-use opportunity costs?

The objective of our research is to extend current conservation applications of modern portfolio theory (MPT) to develop a framework for the cost-efficient budget distribution for a forest carbon payment program that optimizes risk–reward trade-offs in the presence of economic growth uncertainty over time. We consider correlation across space and time of the fluctuating opportunity costs of restoring forestland under changing future economic conditions using a case study of eight states in the central and southern Appalachian region of the United States. The findings suggest that optimal budget allocation decisions that ignore the covariance component of the spatial variance–covariance structure of forest carbon returns fail to minimize the true risk of conservation investment for any level of expected return. The importance of incorporating the spatial covariance in targeting conservation payments is made explicit through alternative approaches using multi-objective (mean–variance) optimization and an ex post analysis with and without the covariance component of the spatial variance–covariance structure of forest carbon return on investment (ROI). A comparison of these approaches against our MPT-based approach revealed misleading risk–return expectations if the ROI covariance is ignored in the spatial targeting of forest carbon payments under uncertainty.

MODELOS DE REGRESSÃO ALEATÓRIA ATRAVÉS DO PESO CORPORAL EM CURVAS DE CRESCIMENTO DE AVES

In modern genetic evaluations, random regression models have been used as a custom tool in order to analyze longitudinal traits such as the ones involved in animal growth. Such traits as body weight have an easy mensuration and an excellent response to selection, which is a suitable and important feature for animal breeding programs. The purpose of this review is to discuss about different random regression models that are involved in farm bird growth. The random regression models are recommended as an alternative to genetic evaluation of traits that are regularly measured during the animal life. These models allow the prediction of regression coefficients that represent the behavior of the additive genetic value for each animal in the specific evaluated trait in relation to time (age). Thus, interminable values for the independent variable are considered 340 Colloquium Agrariae, vol. 13, n. Especial, Jan–Jun, 2017, p. 321-347 ISSN: 1809-8215. DOI: 10.5747/ca.2017.v13.nesp.000238 within a defined interval, through deviations of each animal in relation to an estimated and fixed curve. The covariance component estimates assigned to random regression coefficients allow the covariance estimation between any values of the independent variable for a modeled random effect, which is accomplished by the covariance function. Therefore, random regression models improve the use of the weight information, when covariance structures between the studied ages are taken into account during the evaluation. They also allow the description of the estimated variance components that are involved in growth, besides granting presumptions for others in the curve inside the interval estimation.