Statistical Tests for Joint Analysis of Performance Measures

Recently there has been an increasing interest in the development of new methods using Pareto optimality to deal with multi-objective criteria (for example, accuracy and architectural complexity). Once one has learned a model based on their devised method, the problem is then how to compare it with the state of art. In machine learning, algorithms are typically evaluated by comparing their performance on different data sets by means of statistical tests. Unfortunately, the standard tests used for this purpose are not able to jointly consider performance measures. The aim of this paper is to resolve this issue by developing statistical procedures that are able to account for multiple competing measures at the same time. In particular, we develop two tests: a frequentist procedure based on the generalized likelihood-ratio test and a Bayesian procedure based on a multinomial-Dirichlet conjugate model. We further extend them by discovering conditional independences among measures to reduce the number of parameter of such models, as usually the number of studied cases is very reduced in such comparisons. Real data from a comparison among general purpose classifiers is used to show a practical application of our tests.

Small area estimation for the price of habitation transaction: comparison of different uncertainty measures of temporal EBLUP

Local level planning requires statistics for small areas, but normally due to cost or logistic constraints, sample surveys are often planned to provide reliable estimates only for large geographical regions and large subgroups of a population.

The effect of conservation measures on the spatial and temporal variation of rocky fish assemblages in the Arrábida Marine Park

Marine protected areas (MPAs) have been widely proposed for conservation purposes and as a tool for fisheries management. The Arrábida Marine Park is the first MPA in continental Portugal having a management plan, fully implemented since 2009. The main objective of this study was to evaluate the effect of protection measures on rocky reef fish assemblages and target invertebrates through before-after and control-effect (no-take vs. fished areas) underwater visual surveys and analysis of landings trends. Second, we used surveys before, during and after implementation of the management plan to understand fishers‟ preferences for fishing grounds and adaptation to the new rules, and evaluated the reserve effect through analysis of both ecological responses and fishing effort density. Third, we identified the main oceanographic drivers influencing the structure of reef fish assemblages and predicted the community structure for the last 50 years, in light of climatic change. Overall results suggest positive responses in biomass but not yet in numbers of some commercial species, with no effects on non-target species. The reserve effect is reinforced by the increase in landings of commercial species, despite increased fishing effort density in some areas, especially with octopus traps. Fishing grounds are mainly chosen based on the distribution of target species and associated habitats, but distance to port, weather conditions and safety also influence fishers‟ choices. Moreover, different fisheries respond differently to the protection measures, and within each fishery, individual fishers show distinct strategies, with some operating in a broader area whereas others keep preferred territories. Our results also show that wind stress and temperature are the main oceanographic drivers for rocky reef fish assemblages, with tropicalization of assemblages and polewards movements of species over the last 50 years consistent with temperature trends. We believe this study provides significant lessons for marine conservation and management of coastal systems.

South Carolina Energy Office pre-approved Energy Efficiency Products and Measures

South Carolina law (48-52-640) requires state agencies to submit a disclaimer statement to the State Energy Office with its annual report stating that it did not purchase an energy conservation product that had not been certified by the State Energy Office. This is a list of preapproved products, retrofits and upgrades.

Comparing freight transport strategies and measures in London and Paris

The paper compares the approach being taken to freight transport strategy and the specific policy measures being implemented in London and Paris. It highlights the serious consideration that has been given to freight transport by the Mayors of London and Paris in the last five years. These freight policy considerations are taking place against a background of growing levels of road freight activity, energy use and pollutant emissions in both cities. The key freight transport objectives being followed in London and Paris are similar and focus on improving the efficiency and reliability of freight transport while reducing the negative environmental impacts that it causes. The specific freight transport policy measures being followed show some differences in each city. However, attempts to address problems related to loading and unloading are taking place in both, albeit through different specific initiatives. These policy initiatives have important implications for companies concerned with urban logistics operations.

CO2 reduction and efficiency in road freight traffic: measures, baseline and potential

The objective of our research was to analyse the relevant logistic factors influencing energy efficiency in road freight transport, while quantifying the potential for CO2 reduction. We carried out a survey and linked fuel consumption to transport performance parameters in 50 German haulage companies during 2003. Efficiency ranges from 0.8 tkm to 26 tkm for 1 kg CO2 emissions. The results show a high potential for improvements, given a low level of efficiency in vehicle usage and load factor, scarce use of lightweight vehicle design, incorrectly selected vehicle class and a high proportion of empty runs. Efficiency measures are poorly applied.