Reliability, sufficiency, and the decomposition of proper scores

Scoring rules are an important tool for evaluating the performance of probabilistic forecasting schemes. A scoring rule is called strictly proper if its expectation is optimal if and only if the forecast probability represents the true distribution of the target. In the binary case, strictly proper scoring rules allow for a decomposition into terms related to the resolution and the reliability of a forecast. This fact is particularly well known for the Brier Score. In this article, this result is extended to forecasts for finite-valued targets. Both resolution and reliability are shown to have a positive effect on the score. It is demonstrated that resolution and reliability are directly related to forecast attributes that are desirable on grounds independent of the notion of scores. This finding can be considered an epistemological justification of measuring forecast quality by proper scoring rules. A link is provided to the original work of DeGroot and Fienberg, extending their concepts of sufficiency and refinement. The relation to the conjectured sharpness principle of Gneiting, et al., is elucidated.

Estimating reliability and resolution of probability forecasts through decomposition of the empirical score

References (20)Cited By (1)Export CitationAboutAbstract Proper scoring rules provide a useful means to evaluate probabilistic forecasts. Independent from scoring rules, it has been argued that reliability and resolution are desirable forecast attributes. The mathematical expectation value of the score allows for a decomposition into reliability and resolution related terms, demonstrating a relationship between scoring rules and reliability/resolution. A similar decomposition holds for the empirical (i.e. sample average) score over an archive of forecast–observation pairs. This empirical decomposition though provides a too optimistic estimate of the potential score (i.e. the optimum score which could be obtained through recalibration), showing that a forecast assessment based solely on the empirical resolution and reliability terms will be misleading. The differences between the theoretical and empirical decomposition are investigated, and specific recommendations are given how to obtain better estimators of reliability and resolution in the case of the Brier and Ignorance scoring rule.

On instabilities in data assimilation algorithms

Data assimilation algorithms are a crucial part of operational systems in numerical weather prediction, hydrology and climate science, but are also important for dynamical reconstruction in medical applications and quality control for manufacturing processes. Usually, a variety of diverse measurement data are employed to determine the state of the atmosphere or to a wider system including land and oceans. Modern data assimilation systems use more and more remote sensing data, in particular radiances measured by satellites, radar data and integrated water vapor measurements via GPS/GNSS signals. The inversion of some of these measurements are ill-posed in the classical sense, i.e. the inverse of the operator H which maps the state onto the data is unbounded. In this case, the use of such data can lead to significant instabilities of data assimilation algorithms. The goal of this work is to provide a rigorous mathematical analysis of the instability of well-known data assimilation methods. Here, we will restrict our attention to particular linear systems, in which the instability can be explicitly analyzed. We investigate the three-dimensional variational assimilation and four-dimensional variational assimilation. A theory for the instability is developed using the classical theory of ill-posed problems in a Banach space framework. Further, we demonstrate by numerical examples that instabilities can and will occur, including an example from dynamic magnetic tomography.

Standardized test to evaluate numerical weather prediction algorithms

In order to assist in comparing the computational techniques used in different models, the authors propose a standardized set of one-dimensional numerical experiments that could be completed for each model. The results of these experiments, with a simplified form of the computational representation for advection, diffusion, pressure gradient term, Coriolis term, and filter used in the models, should be reported in the peer-reviewed literature. Specific recommendations are described in this paper.

System identification algorithms for the analysis of dielectric responses from broadband spectroscopies

We discuss the modeling of dielectric responses for an electromagnetically excited network of capacitors and resistors using a systems identification framework. Standard models that assume integral order dynamics are augmented to incorporate fractional order dynamics. This enables us to relate more faithfully the modeled responses to those reported in the Dielectrics literature.

Reliability of regional climate model trends

A necessary condition for a good probabilistic forecast is that the forecast system is shown to be reliable: forecast probabilities should equal observed probabilities verified over a large number of cases. As climate change trends are now emerging from the natural variability, we can apply this concept to climate predictions and compute the reliability of simulated local and regional temperature and precipitation trends (1950–2011) in a recent multi-model ensemble of climate model simulations prepared for the Intergovernmental Panel on Climate Change (IPCC) fifth assessment report (AR5). With only a single verification time, the verification is over the spatial dimension. The local temperature trends appear to be reliable. However, when the global mean climate response is factored out, the ensemble is overconfident: the observed trend is outside the range of modelled trends in many more regions than would be expected by the model estimate of natural variability and model spread. Precipitation trends are overconfident for all trend definitions. This implies that for near-term local climate forecasts the CMIP5 ensemble cannot simply be used as a reliable probabilistic forecast.

Mobility based energy efficient and multi-sink algorithms for consumer home networks

With the fast development of the Internet, wireless communications and semiconductor devices, home networking has received significant attention. Consumer products can collect and transmit various types of data in the home environment. Typical consumer sensors are often equipped with tiny, irreplaceable batteries and it therefore of the utmost importance to design energy efficient algorithms to prolong the home network lifetime and reduce devices going to landfill. Sink mobility is an important technique to improve home network performance including energy consumption, lifetime and end-to-end delay. Also, it can largely mitigate the hot spots near the sink node. The selection of optimal moving trajectory for sink node(s) is an NP-hard problem jointly optimizing routing algorithms with the mobile sink moving strategy is a significant and challenging research issue. The influence of multiple static sink nodes on energy consumption under different scale networks is first studied and an Energy-efficient Multi-sink Clustering Algorithm (EMCA) is proposed and tested. Then, the influence of mobile sink velocity, position and number on network performance is studied and a Mobile-sink based Energy-efficient Clustering Algorithm (MECA) is proposed. Simulation results validate the performance of the proposed two algorithms which can be deployed in a consumer home network environment.

IMILAST: a community effort to intercompare extratropical cyclone detection and tracking algorithms

The variability of results from different automated methods of detection and tracking of extratropical cyclones is assessed in order to identify uncertainties related to the choice of method. Fifteen international teams applied their own algorithms to the same dataset—the period 1989–2009 of interim European Centre for Medium-Range Weather Forecasts (ECMWF) Re-Analysis (ERAInterim) data. This experiment is part of the community project Intercomparison of Mid Latitude Storm Diagnostics (IMILAST; see www.proclim.ch/imilast/index.html). The spread of results for cyclone frequency, intensity, life cycle, and track location is presented to illustrate the impact of using different methods. Globally, methods agree well for geographical distribution in large oceanic regions, interannual variability of cyclone numbers, geographical patterns of strong trends, and distribution shape for many life cycle characteristics. In contrast, the largest disparities exist for the total numbers of cyclones, the detection of weak cyclones, and distribution in some densely populated regions. Consistency between methods is better for strong cyclones than for shallow ones. Two case studies of relatively large, intense cyclones reveal that the identification of the most intense part of the life cycle of these events is robust between methods, but considerable differences exist during the development and the dissolution phases.

Graph-based algorithms for comparison and prediction of household-level energy use profiles

We present an efficient graph-based algorithm for quantifying the similarity of household-level energy use profiles, using a notion of similarity that allows for small time–shifts when comparing profiles. Experimental results on a real smart meter data set demonstrate that in cases of practical interest our technique is far faster than the existing method for computing the same similarity measure. Having a fast algorithm for measuring profile similarity improves the efficiency of tasks such as clustering of customers and cross-validation of forecasting methods using historical data. Furthermore, we apply a generalisation of our algorithm to produce substantially better household-level energy use forecasts from historical smart meter data.

Terrestrial ecosystems from space: a review of earth observation products for macroecology applications

Aim Earth observation (EO) products are a valuable alternative to spectral vegetation indices. We discuss the availability of EO products for analysing patterns in macroecology, particularly related to vegetation, on a range of spatial and temporal scales. Location Global. Methods We discuss four groups of EO products: land cover/cover change, vegetation structure and ecosystem productivity, fire detection, and digital elevation models. We address important practical issues arising from their use, such as assumptions underlying product generation, product accuracy and product transferability between spatial scales. We investigate the potential of EO products for analysing terrestrial ecosystems. Results Land cover, productivity and fire products are generated from long-term data using standardized algorithms to improve reliability in detecting change of land surfaces. Their global coverage renders them useful for macroecology. Their spatial resolution (e.g. GLOBCOVER vegetation, 300 m; MODIS vegetation and fire, ≥ 500 m; ASTER digital elevation, 30 m) can be a limiting factor. Canopy structure and productivity products are based on physical approaches and thus are independent of biome-specific calibrations. Active fire locations are provided in near-real time, while burnt area products show actual area burnt by fire. EO products can be assimilated into ecosystem models, and their validation information can be employed to calculate uncertainties during subsequent modelling. Main conclusions Owing to their global coverage and long-term continuity, EO end products can significantly advance the field of macroecology. EO products allow analyses of spatial biodiversity, seasonal dynamics of biomass and productivity, and consequences of disturbances on regional to global scales. Remaining drawbacks include inter-operability between products from different sensors and accuracy issues due to differences between assumptions and models underlying the generation of different EO products. Our review explains the nature of EO products and how they relate to particular ecological variables across scales to encourage their wider use in ecological applications.

The end of the opportunism vs trust debate: bounded reliability as a new envelope concept in research on MNE governance

Modern transaction cost economics (TCE) thinking has developed into a key intellectual foundation of international business (IB) research, but the Williamsonian version has faced substantial criticism for adopting the behavioral assumption of opportunism. In this paper we assess both the opportunism concept and existing alternatives such as trust within the context of IB research, especially work on multinational enterprise (MNE) governance. Case analyses of nine global MNEs illustrate an alternative to the opportunism assumption that captures more fully the mechanisms underlying failed commitments inside the MNE. As a substitute for the often-criticized assumption of opportunism, we propose the envelope concept of bounded reliability (BRel), an assumption that represents more accurately and more completely the reasons for failed commitments, without invalidating the other critical assumption in conventional TCE (and internalization theory) thinking, namely the widely accepted envelope concept of bounded rationality (BRat). Bounded reliability as an envelope concept includes two main components, within the context of global MNE management: opportunism as intentional deceit, and benevolent preference reversal. The implications for IB research of adopting the bounded reliability concept are far reaching, as this concept may increase the legitimacy of comparative institutional analysis in the social sciences.

Examining reliability of seasonal to decadal sea surface temperature forecasts: the role of ensemble dispersion

Useful probabilistic climate forecasts on decadal timescales should be reliable (i.e. forecast probabilities match the observed relative frequencies) but this is seldom examined. This paper assesses a necessary condition for reliability, that the ratio of ensemble spread to forecast error being close to one, for seasonal to decadal sea surface temperature retrospective forecasts from the Met Office Decadal Prediction System (DePreSys). Factors which may affect reliability are diagnosed by comparing this spread-error ratio for an initial condition ensemble and two perturbed physics ensembles for initialized and uninitialized predictions. At lead times less than 2 years, the initialized ensembles tend to be under-dispersed, and hence produce overconfident and hence unreliable forecasts. For longer lead times, all three ensembles are predominantly over-dispersed. Such over-dispersion is primarily related to excessive inter-annual variability in the climate model. These findings highlight the need to carefully evaluate simulated variability in seasonal and decadal prediction systems.Useful probabilistic climate forecasts on decadal timescales should be reliable (i.e. forecast probabilities match the observed relative frequencies) but this is seldom examined. This paper assesses a necessary condition for reliability, that the ratio of ensemble spread to forecast error being close to one, for seasonal to decadal sea surface temperature retrospective forecasts from the Met Office Decadal Prediction System (DePreSys). Factors which may affect reliability are diagnosed by comparing this spread-error ratio for an initial condition ensemble and two perturbed physics ensembles for initialized and uninitialized predictions. At lead times less than 2 years, the initialized ensembles tend to be under-dispersed, and hence produce overconfident and hence unreliable forecasts. For longer lead times, all three ensembles are predominantly over-dispersed. Such over-dispersion is primarily related to excessive inter-annual variability in the climate model. These findings highlight the need to carefully evaluate simulated variability in seasonal and decadal prediction systems.

Examining reliability of seasonal to decadal sea surface temperature forecasts: the role of ensemble dispersion

Useful probabilistic climate forecasts on decadal timescales should be reliable (i.e. forecast probabilities match the observed relative frequencies) but this is seldom examined. This paper assesses a necessary condition for reliability, that the ratio of ensemble spread to forecast error being close to one, for seasonal to decadal sea surface temperature retrospective forecasts from the Met Office Decadal Prediction System (DePreSys). Factors which may affect reliability are diagnosed by comparing this spread-error ratio for an initial condition ensemble and two perturbed physics ensembles for initialized and uninitialized predictions. At lead times less than 2 years, the initialized ensembles tend to be under-dispersed, and hence produce overconfident and hence unreliable forecasts. For longer lead times, all three ensembles are predominantly over-dispersed. Such over-dispersion is primarily related to excessive inter-annual variability in the climate model. These findings highlight the need to carefully evaluate simulated variability in seasonal and decadal prediction systems.Useful probabilistic climate forecasts on decadal timescales should be reliable (i.e. forecast probabilities match the observed relative frequencies) but this is seldom examined. This paper assesses a necessary condition for reliability, that the ratio of ensemble spread to forecast error being close to one, for seasonal to decadal sea surface temperature retrospective forecasts from the Met Office Decadal Prediction System (DePreSys). Factors which may affect reliability are diagnosed by comparing this spread-error ratio for an initial condition ensemble and two perturbed physics ensembles for initialized and uninitialized predictions. At lead times less than 2 years, the initialized ensembles tend to be under-dispersed, and hence produce overconfident and hence unreliable forecasts. For longer lead times, all three ensembles are predominantly over-dispersed. Such over-dispersion is primarily related to excessive inter-annual variability in the climate model. These findings highlight the need to carefully evaluate simulated variability in seasonal and decadal prediction systems.

Sea surface temperature climate change initiative: alternative image classification algorithms for sea-ice affected oceans

We present a Bayesian image classification scheme for discriminating cloud, clear and sea-ice observations at high latitudes to improve identification of areas of clear-sky over ice-free ocean for SST retrieval. We validate the image classification against a manually classified dataset using Advanced Along Track Scanning Radiometer (AATSR) data. A three way classification scheme using a near-infrared textural feature improves classifier accuracy by 9.9 % over the nadir only version of the cloud clearing used in the ATSR Reprocessing for Climate (ARC) project in high latitude regions. The three way classification gives similar numbers of cloud and ice scenes misclassified as clear but significantly more clear-sky cases are correctly identified (89.9 % compared with 65 % for ARC). We also demonstrate the poetential of a Bayesian image classifier including information from the 0.6 micron channel to be used in sea-ice extent and ice surface temperature retrieval with 77.7 % of ice scenes correctly identified and an overall classifier accuracy of 96 %.