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.

Skilloscopy: Bayesian modeling of decision makers' skill

This paper proposes and demonstrates an approach, Skilloscopy, to the assessment of decision makers. In an increasingly sophisticated, connected and information-rich world, decision making is becoming both more important and more difficult. At the same time, modelling decision-making on computers is becoming more feasible and of interest, partly because the information-input to those decisions is increasingly on record. The aims of Skilloscopy are to rate and rank decision makers in a domain relative to each other: the aims do not include an analysis of why a decision is wrong or suboptimal, nor the modelling of the underlying cognitive process of making the decisions. In the proposed method a decision-maker is characterised by a probability distribution of their competence in choosing among quantifiable alternatives. This probability distribution is derived by classic Bayesian inference from a combination of prior belief and the evidence of the decisions. Thus, decision-makers’ skills may be better compared, rated and ranked. The proposed method is applied and evaluated in the gamedomain of Chess. A large set of games by players across a broad range of the World Chess Federation (FIDE) Elo ratings has been used to infer the distribution of players’ rating directly from the moves they play rather than from game outcomes. Demonstration applications address questions frequently asked by the Chess community regarding the stability of the Elo rating scale, the comparison of players of different eras and/or leagues, and controversial incidents possibly involving fraud. The method of Skilloscopy may be applied in any decision domain where the value of the decision-options can be quantified.

Probability of survival in a random exchange economy with dependent agents

In this paper I analyze the general equilibrium in a random Walrasian economy. Dependence among agents is introduced in the form of dependency neighborhoods. Under the uncertainty, an agent may fail to survive due to a meager endowment in a particular state (direct effect), as well as due to unfavorable equilibrium price system at which the value of the endowment falls short of the minimum needed for survival (indirect terms-of-trade effect). To illustrate the main result I compute the stochastic limit of equilibrium price and probability of survival of an agent in a large Cobb-Douglas economy.

Sparse probability density function estimation using the minimum integrated square error

We develop a new sparse kernel density estimator using a forward constrained regression framework, within which the nonnegative and summing-to-unity constraints of the mixing weights can easily be satisfied. Our main contribution is to derive a recursive algorithm to select significant kernels one at time based on the minimum integrated square error (MISE) criterion for both the selection of kernels and the estimation of mixing weights. The proposed approach is simple to implement and the associated computational cost is very low. Specifically, the complexity of our algorithm is in the order of the number of training data N, which is much lower than the order of N2 offered by the best existing sparse kernel density estimators. Numerical examples are employed to demonstrate that the proposed approach is effective in constructing sparse kernel density estimators with comparable accuracy to those of the classical Parzen window estimate and other existing sparse kernel density estimators.

Rossby wave-breaking analysis of explosive cyclones in the Euro-Atlantic sector

The two-way relationship between Rossby Wave-Breaking (RWB) and intensification of extra tropical cyclones is analysed over the Euro-Atlantic sector. In particular, the timing, intensity and location of cyclone development are related to RWB occurrences. For this purpose, two potential-temperature based indices are used to detect and classify anticyclonic and cyclonic RWB episodes from ERA-40 Re-Analysis data. Results show that explosive cyclogenesis over the North Atlantic (NA) is fostered by enhanced occurrence of RWB on days prior to the cyclone’s maximum intensification. Under such conditions, the eddy-driven jet stream is accelerated over the NA, thus enhancing conditions for cyclogenesis. For explosive cyclogenesis over the eastern NA, enhanced cyclonic RWB over eastern Greenland and anticyclonic RWB over the sub-tropical NA are observed. Typically only one of these is present in any given case, with the RWB over eastern Greenland being more frequent than its southern counterpart. This leads to an intensification of the jet over the eastern NA and enhanced probability of windstorms reaching Western Europe. Explosive cyclones evolving under simultaneous RWB on both sides of the jet feature a higher mean intensity and deepening rates than cyclones preceded by a single RWB event. Explosive developments over the western NA are typically linked to a single area of enhanced cyclonic RWB over western Greenland. Here, the eddy-driven jet is accelerated over the western NA. Enhanced occurrence of cyclonic RWB over southern Greenland and anticyclonic RWB over Europe is also observed after explosive cyclogenesis, potentially leading to the onset of Scandinavian Blocking. However, only very intense developments have a considerable influence on the large-scale atmospheric flow. Non-explosive cyclones depict no sign of enhanced RWB over the whole NA area. We conclude that the links between RWB and cyclogenesis over the Euro-Atlantic sector are sensitive to the cyclone’s maximum intensity, deepening rate and location.

Probabilistic physically-based cloud screening of satellite infra-red imagery for operational sea surface temperature retrieval

We propose and demonstrate a fully probabilistic (Bayesian) approach to the detection of cloudy pixels in thermal infrared (TIR) imagery observed from satellite over oceans. Using this approach, we show how to exploit the prior information and the fast forward modelling capability that are typically available in the operational context to obtain improved cloud detection. The probability of clear sky for each pixel is estimated by applying Bayes' theorem, and we describe how to apply Bayes' theorem to this problem in general terms. Joint probability density functions (PDFs) of the observations in the TIR channels are needed; the PDFs for clear conditions are calculable from forward modelling and those for cloudy conditions have been obtained empirically. Using analysis fields from numerical weather prediction as prior information, we apply the approach to imagery representative of imagers on polar-orbiting platforms. In comparison with the established cloud-screening scheme, the new technique decreases both the rate of failure to detect cloud contamination and the false-alarm rate by one quarter. The rate of occurrence of cloud-screening-related errors of >1 K in area-averaged SSTs is reduced by 83%. Copyright © 2005 Royal Meteorological Society.

Forecast encompassing tests and probability forecasts

We consider tests of forecast encompassing for probability forecasts, for both quadratic and logarithmic scoring rules. We propose test statistics for the null of forecast encompassing, present the limiting distributions of the test statistics, and investigate the impact of estimating the forecasting models' parameters on these distributions. The small-sample performance is investigated, in terms of small numbers of forecasts and model estimation sample sizes. We show the usefulness of the tests for the evaluation of recession probability forecasts from logit models with different leading indicators as explanatory variables, and for evaluating survey-based probability forecasts.

A fast algorithm for sparse probability density function construction

A new sparse kernel density estimator is introduced. Our main contribution is to develop a recursive algorithm for the selection of significant kernels one at time using the minimum integrated square error (MISE) criterion for both kernel selection. The proposed approach is simple to implement and the associated computational cost is very low. Numerical examples are employed to demonstrate that the proposed approach is effective in constructing sparse kernel density estimators with competitive accuracy to existing kernel density estimators.

On the transfer of prior tests or study events to subsequent study

Tests, as learning events, are often more effective than are additional study opportunities, especially when recall is tested after a long retention interval. To what degree, though, do prior test or study events support subsequent study activities? We set out to test an implication of Bjork and Bjork’s (1992) new theory of disuse—that, under some circumstances, prior study may facilitate subsequent study more than does prior testing. Participants learned English–Swahili translations and then underwent a practice phase during which some items were tested (without feedback) and other items were restudied. Although tested items were better recalled after a 1-week delay than were restudied items, this benefit did not persist after participants had the opportunity to study the items again via feedback. In fact, after this additional study opportunity, items that had been restudied earlier were better recalled than were items that had been tested earlier. These results suggest that measuring the memorial consequences of testing requires more than a single test of retention and, theoretically, a consideration of the differing status of initially recallable and nonrecallable items.

Leader of the pack? German monetary dominance in Europe prior to EMU

In this paper, the monetary policy independence of European nations in the years before European Economic and Monetary Union (EMU) is investigated using cointegration techniques. Daily data is used to assess pairwise relationships between individual EMU nations and ‘lead’ nation Germany, to assess the hypothesis that Germany was the dominant European nation prior to EMU. By and large our econometric investigations support this hypothesis, and lead us to conclude that the only European nation to lose monetary policy independence in the light of monetary union was Germany. Our results have important policy implications. Given that the loss of monetary policy independence is generally viewed as the main cost of monetary unification, our findings suggest a reconsideration of the costs and benefits of monetary integration. A country can only lose what it has, and in Europe the countries that joined EMU — spare Germany — apparently did not have much to lose, at least not in terms of monetary independence. Instead, they actually gained monetary policy influence by getting a seat in the ECB's governing council which is responsible for setting interest policy in the euro area.

Evaluating the Survey of Professional Forecasters probability distributions of expected inflation based on derived event probability forecasts

Techniques are proposed for evaluating forecast probabilities of events. The tools are especially useful when, as in the case of the Survey of Professional Forecasters (SPF) expected probability distributions of inflation, recourse cannot be made to the method of construction in the evaluation of the forecasts. The tests of efficiency and conditional efficiency are applied to the forecast probabilities of events of interest derived from the SPF distributions, and supplement a whole-density evaluation of the SPF distributions based on the probability integral transform approach.

Combining probability forecasts

We consider different methods for combining probability forecasts. In empirical exercises, the data generating process of the forecasts and the event being forecast is not known, and therefore the optimal form of combination will also be unknown. We consider the properties of various combination schemes for a number of plausible data generating processes, and indicate which types of combinations are likely to be useful. We also show that whether forecast encompassing is found to hold between two rival sets of forecasts or not may depend on the type of combination adopted. The relative performances of the different combination methods are illustrated, with an application to predicting recession probabilities using leading indicators.

Probability distributions or point predictions? Survey forecasts of US output growth and inflation

We consider whether survey respondents’ probability distributions, reported as histograms, provide reliable and coherent point predictions, when viewed through the lens of a Bayesian learning model. We argue that a role remains for eliciting directly-reported point predictions in surveys of professional forecasters.