Parameter estimation based on stacked regression and evolutionary algorithms

A new parameter-estimation algorithm, which minimises the cross-validated prediction error for linear-in-the-parameter models, is proposed, based on stacked regression and an evolutionary algorithm. It is initially shown that cross-validation is very important for prediction in linear-in-the-parameter models using a criterion called the mean dispersion error (MDE). Stacked regression, which can be regarded as a sophisticated type of cross-validation, is then introduced based on an evolutionary algorithm, to produce a new parameter-estimation algorithm, which preserves the parsimony of a concise model structure that is determined using the forward orthogonal least-squares (OLS) algorithm. The PRESS prediction errors are used for cross-validation, and the sunspot and Canadian lynx time series are used to demonstrate the new algorithms.

Application of logistic regression for fault analysis in an industrial printing process

A statistical technique for fault analysis in industrial printing is reported. The method specifically deals with binary data, for which the results of the production process fall into two categories, rejected or accepted. The method is referred to as logistic regression, and is capable of predicting future fault occurrences by the analysis of current measurements from machine parts sensors. Individual analysis of each type of fault can determine which parts of the plant have a significant influence on the occurrence of such faults; it is also possible to infer which measurable process parameters have no significant influence on the generation of these faults. Information derived from the analysis can be helpful in the operator's interpretation of the current state of the plant. Appropriate actions may then be taken to prevent potential faults from occurring. The algorithm is being implemented as part of an applied self-learning expert system.

Societal choice for climate change futures: trees, biotechnology and clean development

The rate and magnitude of predicted climate change require that we urgently mitigate emissions or sequester carbon on a substantial scale in order to avoid runaway climate change. Geo- and bioengineering solutions are increasingly proposed as viable and practical strategies for tackling global warming. Biotechnology companies are already developing transgenic “super carbon-absorbing” trees, which are sold as a cost-effective and relatively low-risk means of sequestering carbon. The question posed in this article is, Do super carbon trees provide real benefits or are they merely a fanciful illusion? It remains unclear whether growing these trees makes sense in terms of the carbon cost of production and the actual storage of carbon. In particular, it is widely acknowledged that “carbon-eating” trees fail to sequester as much carbon as they oxidize and return to the atmosphere; moreover, there are concerns about the biodiversity impacts of large-scale monoculture plantations. The potential social and ecological risks and opportunities presented by such controversial solutions warrant a societal dialogue.

Biomass functions and expansion factors in young Norway spruce (Picea abies [L.] Karst) trees

Roots, stems, branches and needles of 160 Norway spruce trees younger than 10 years were sampled in seven forest stands in central Slovakia in order to establish their biomassfunctions (BFs) and biomassexpansionfactors (BEFs). We tested three models for each biomass pool based on the stem base diameter, tree height and the two parameters combined. BEF values decreased for all spruce components with increasing height and diameter, which was most evident in very young trees under 1 m in height. In older trees, the values of BEFs did tend to stabilise at the height of 3–4 m. We subsequently used the BEFs to calculate dry biomass of the stands based on average stem base diameter and tree height. Total stand biomass grew with increasing age of the stands from about 1.0 Mg ha−1 at 1.5 years to 44.3 Mg ha−1 at 9.5 years. The proportion of stem and branch biomass was found to increase with age, while that of needles was fairly constant and the proportion of root biomass did decrease as the stands grew older.

Who benefits from reducing the cost of formality? Quantile regression discontinuity analysis

This paper studies the effects of increasing formality via tax reduction and simplification schemes on micro-firm performance. It uses the 1997 Brazilian SIMPLES program. We develop a simple theoretical model to show that SIMPLES has an impact only on a segment of the micro-firm population, for which the effect of formality on firm performance can be identified, and that can be analyzed along the single dimensional quantiles of the conditional firm revenues. To estimate the effect of formality, we use an econometric approach that compares eligible and non-eligible firms, born before and after SIMPLES in a local interval about the introduction of SIMPLES. We use an estimator that combines both quantile regression and the regression discontinuity identification strategy. The empirical results corroborate the positive effect of formality on microfirms' performance and produce a clear characterization of who benefits from these programs.

The trees and the bees: using enforcement and income projects to protect forests and rural livelihoods through spatial joint production

Forest managers in developing countries enforce extraction restrictions to limit forest degradation. In response, villagers may displace some of their extraction to other forests, which generates “leakage” of degradation. Managers also implement poverty alleviation projects to compensate for lost resource access or to induce conservation. We develop a model of spatial joint production of bees and fuelwood that is based on forest-compatible projects such as beekeeping in Thailand, Tanzania, and Mexico. We demonstrate that managers can better determine the amount and pattern of degradation by choosing the location of both enforcement and the forest-based activity.

An elastic net orthogonal forward regression algorithm

In this paper we propose an efficient two-level model identification method for a large class of linear-in-the-parameters models from the observational data. A new elastic net orthogonal forward regression (ENOFR) algorithm is employed at the lower level to carry out simultaneous model selection and elastic net parameter estimation. The two regularization parameters in the elastic net are optimized using a particle swarm optimization (PSO) algorithm at the upper level by minimizing the leave one out (LOO) mean square error (LOOMSE). Illustrative examples are included to demonstrate the effectiveness of the new approaches.

Distributed hoeffding trees for pocket data mining

Collaborative mining of distributed data streams in a mobile computing environment is referred to as Pocket Data Mining PDM. Hoeffding trees techniques have been experimentally and analytically validated for data stream classification. In this paper, we have proposed, developed and evaluated the adoption of distributed Hoeffding trees for classifying streaming data in PDM applications. We have identified a realistic scenario in which different users equipped with smart mobile devices run a local Hoeffding tree classifier on a subset of the attributes. Thus, we have investigated the mining of vertically partitioned datasets with possible overlap of attributes, which is the more likely case. Our experimental results have validated the efficiency of our proposed model achieving promising accuracy for real deployment.

Thorn-dwelling ants provide antiherbivore defence for camelthorn trees, Vachellia erioloba, in Namibia

Ants are widely employed by plants as an antiherbivore defence. A single host plant can associate with multiple, symbiotic ant species, although usually only a single ant species at a time. Different plant-ant species may vary in the degree to which they defend their host plant. In Kenya, ant–acacia interactions are well studied, but less is known about systems elsewhere in Africa. A southern African species, Vachellia erioloba, is occupied by thorn-dwelling ants from three different genera. Unusually, multiple colonies of all these ants simultaneously and stably inhabit trees. We investigated if the ants on V. erioloba (i) deter insect herbivores; (ii) differ in their effectiveness depending on the identity of the herbivore; and (iii) protect the tree against an important herbivore, the larvae of the lepidopteran Gonometa postica. We show that experimental exclusion of ants leads to greater levels of herbivory on trees. The ants inhabiting V. erioloba are an effective deterrent against hemipteran and coleopteran, but not lepidopteran herbivores. Defensive services do not vary among ant species, but only Crematogaster ants exhibit aggression towards G. postica. This highlights the potential of the V. erioloba–ant mutualism for studying ant–plant interactions that involve multiple, simultaneously resident thorn-dwelling ant species.