Likelihood-free estimation of model evidence

Statistical methods of inference typically require the likelihood function to be computable in a reasonable amount of time. The class of “likelihood-free” methods termed Approximate Bayesian Computation (ABC) is able to eliminate this requirement, replacing the evaluation of the likelihood with simulation from it. Likelihood-free methods have gained in efficiency and popularity in the past few years, following their integration with Markov Chain Monte Carlo (MCMC) and Sequential Monte Carlo (SMC) in order to better explore the parameter space. They have been applied primarily to estimating the parameters of a given model, but can also be used to compare models. Here we present novel likelihood-free approaches to model comparison, based upon the independent estimation of the evidence of each model under study. Key advantages of these approaches over previous techniques are that they allow the exploitation of MCMC or SMC algorithms for exploring the parameter space, and that they do not require a sampler able to mix between models. We validate the proposed methods using a simple exponential family problem before providing a realistic problem from human population genetics: the comparison of different demographic models based upon genetic data from the Y chromosome.

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.

Corporate responsibility, supply chain partnership and performance: An empirical examination

Unlike corporate and business levels, there is little research examining corporate responsibility (CR) at the functional level of the firm including supply chain strategy. The results of a firm-level survey show that CR internal awareness, and monitoring CR performance are positively related to the supply chain partnership approach, however sharing CR best practices is negatively associated. Furthermore, the impact of CR on firm performance is mediated by the functional behaviour of supply chain partnership formation. Our study provides support for including CR awareness building and monitoring in the development of partnerships but cautions against imposing CR best practices on suppliers.

Futures basis, inventory and commodity price volatility: an empirical analysis

We employ a large dataset of physical inventory data on 21 different commodities for the period 1993–2011 to empirically analyze the behavior of commodity prices and their volatility as predicted by the theory of storage. We examine two main issues. First, we analyze the relationship between inventory and the shape of the forward curve. Low (high) inventory is associated with forward curves in backwardation (contango), as the theory of storage predicts. Second, we show that price volatility is a decreasing function of inventory for the majority of commodities in our sample. This effect is more pronounced in backwardated markets. Our findings are robust with respect to alternative inventory measures and over the recent commodity price boom.

Online data processing consent under EU law: a theoretical framework and empirical evidence from the UK

This article analyses the results of an empirical study on the 200 most popular UK-based websites in various sectors of e-commerce services. The study provides empirical evidence on unlawful processing of personal data. It comprises a survey on the methods used to seek and obtain consent to process personal data for direct marketing and advertisement, and a test on the frequency of unsolicited commercial emails (UCE) received by customers as a consequence of their registration and submission of personal information to a website. Part One of the article presents a conceptual and normative account of data protection, with a discussion of the ethical values on which EU data protection law is grounded and an outline of the elements that must be in place to seek and obtain valid consent to process personal data. Part Two discusses the outcomes of the empirical study, which unveils a significant departure between EU legal theory and practice in data protection. Although a wide majority of the websites in the sample (69%) has in place a system to ask separate consent for engaging in marketing activities, it is only 16.2% of them that obtain a consent which is valid under the standards set by EU law. The test with UCE shows that only one out of three websites (30.5%) respects the will of the data subject not to receive commercial communications. It also shows that, when submitting personal data in online transactions, there is a high probability (50%) of incurring in a website that will ignore the refusal of consent and will send UCE. The article concludes that there is severe lack of compliance of UK online service providers with essential requirements of data protection law. In this respect, it suggests that there is inappropriate standard of implementation, information and supervision by the UK authorities, especially in light of the clarifications provided at EU level.

Handbook of research methods and applications in empirical finance

The objective of this book is to present the quantitative techniques that are commonly employed in empirical finance research together with real world, state of the art research examples. Each chapter is written by international experts in their fields. The unique approach is to describe a question or issue in finance and then to demonstrate the methodologies that may be used to solve it. All of the techniques described are used to address real problems rather than being presented for their own sake and the areas of application have been carefully selected so that a broad range of methodological approaches can be covered. This book is aimed primarily at doctoral researchers and academics who are engaged in conducting original empirical research in finance. In addition, the book will be useful to researchers in the financial markets and also advanced Masters-level students who are writing dissertations.

Empirical test of an agricultural landscape model: the importance of farmer preference for risk aversion and crop complexity

Developing models to predict the effects of social and economic change on agricultural landscapes is an important challenge. Model development often involves making decisions about which aspects of the system require detailed description and which are reasonably insensitive to the assumptions. However, important components of the system are often left out because parameter estimates are unavailable. In particular, measurements of the relative influence of different objectives, such as risk, environmental management, on farmer decision making, have proven difficult to quantify. We describe a model that can make predictions of land use on the basis of profit alone or with the inclusion of explicit additional objectives. Importantly, our model is specifically designed to use parameter estimates for additional objectives obtained via farmer interviews. By statistically comparing the outputs of this model with a large farm-level land-use data set, we show that cropping patterns in the United Kingdom contain a significant contribution from farmer’s preference for objectives other than profit. In particular, we found that risk aversion had an effect on the accuracy of model predictions, whereas preference for a particular number of crops grown was less important. While nonprofit objectives have frequently been identified as factors in farmers’ decision making, our results take this analysis further by demonstrating the relationship between these preferences and actual cropping patterns.

Can adherence to dietary guidelines address excess caloric intake? An empirical assessment for the UK

The facilitation of healthier dietary choices by consumers is a key element of government strategies to combat the rising incidence of obesity in developed and developing countries. Public health campaigns to promote healthier eating often target compliance with recommended dietary guidelines for consumption of individual nutrients such as fats and added sugars. This paper examines the association between improved compliance with dietary guidelines for individual nutrients and excess calorie intake, the most proximate determinant of obesity risk. We apply quantile regressions and counterfactual decompositions to cross-sectional data from the National Diet and Nutrition Survey (2000-01) to assess how excess calorie consumption patterns in the UK are likely to change with improved compliance with dietary guidelines. We find that the effects of compliance vary significantly across different quantiles of calorie consumption. Our results show that compliance with dietary guidelines for individual nutrients, even if successfully achieved, is likely to be associated with only modest shifts in excess calorie consumption patterns. Consequently, public health campaigns that target compliance with dietary guidelines for specific nutrients in isolation are unlikely to have a significant effect on the obesity risk faced by the population.

The interaction of volatility, volume and skewness: empirical evidence from REITs

This paper considers how trading volume impacts upon the first three moments of REIT returns. Consistent with previous studies of the broader stock market, we find that volume is a significant factor with respect to both returns and volatility. We also find evidence supportive of the Hong & Stein’s (2003) Investor Heterogeneity Theory with respect to the finding that skewness in REIT index returns is significantly related to volume. Furthermore, we also report findings that show the influence of the variability of volume with skewness.

The Empirical Forcing Function as a tool for the diagnosis of large-scale atmospheric anomalies

The time-mean quasi-geostrophic potential vorticity equation of the atmospheric flow on isobaric surfaces can explicitly include an atmospheric (internal) forcing term of the stationary-eddy flow. In fact, neglecting some non-linear terms in this equation, this forcing can be mathematically expressed as a single function, called Empirical Forcing Function (EFF), which is equal to the material derivative of the time-mean potential vorticity. Furthermore, the EFF can be decomposed as a sum of seven components, each one representing a forcing mechanism of different nature. These mechanisms include diabatic components associated with the radiative forcing, latent heat release and frictional dissipation, and components related to transient eddy transports of heat and momentum. All these factors quantify the role of the transient eddies in forcing the atmospheric circulation. In order to assess the relevance of the EFF in diagnosing large-scale anomalies in the atmospheric circulation, the relationship between the EFF and the occurrence of strong North Atlantic ridges over the Eastern North Atlantic is analyzed, which are often precursors of severe droughts over Western Iberia. For such events, the EFF pattern depicts a clear dipolar structure over the North Atlantic; cyclonic (anticyclonic) forcing of potential vorticity is found upstream (downstream) of the anomalously strong ridges. Results also show that the most significant components are related to the diabatic processes. Lastly, these results highlight the relevance of the EFF in diagnosing large-scale anomalies, also providing some insight into their interaction with different physical mechanisms.

Constructing summary statistics for approximate Bayesian computation: semi-automatic approximate Bayesian computation

Many modern statistical applications involve inference for complex stochastic models, where it is easy to simulate from the models, but impossible to calculate likelihoods. Approximate Bayesian computation (ABC) is a method of inference for such models. It replaces calculation of the likelihood by a step which involves simulating artificial data for different parameter values, and comparing summary statistics of the simulated data with summary statistics of the observed data. Here we show how to construct appropriate summary statistics for ABC in a semi-automatic manner. We aim for summary statistics which will enable inference about certain parameters of interest to be as accurate as possible. Theoretical results show that optimal summary statistics are the posterior means of the parameters. Although these cannot be calculated analytically, we use an extra stage of simulation to estimate how the posterior means vary as a function of the data; and we then use these estimates of our summary statistics within ABC. Empirical results show that our approach is a robust method for choosing summary statistics that can result in substantially more accurate ABC analyses than the ad hoc choices of summary statistics that have been proposed in the literature. We also demonstrate advantages over two alternative methods of simulation-based inference.

Updated empirical model of genetically-modified maize grain production practices to achieve European Union labeling thresholds

An updated empirical approach is proposed for specifying coexistence requirements for genetically modified (GM) maize (Zea mays L.) production to ensure compliance with the 0.9% labeling threshold for food and feed in the European Union. The model improves on a previously published (Gustafson et al., 2006) empirical model by adding recent data sources to supplement the original database and including the following additional cases: (i) more than one GM maize source field adjacent to the conventional or organic field, (ii) the possibility of so-called “stacked” varieties with more than one GM trait, and (iii) lower pollen shed in the non-GM receptor field. These additional factors lead to the possibility for somewhat wider combinations of isolation distance and border rows than required in the original version of the empirical model. For instance, in the very conservative case of a 1-ha square non-GM maize field surrounded on all four sides by homozygous GM maize with 12 m isolation (the effective isolation distance for a single GM field), non-GM border rows of 12 m are required to be 95% confident of gene flow less than 0.9% in the non-GM field (with adventitious presence of 0.3%). Stacked traits of higher GM mass fraction and receptor fields of lower pollen shed would require a greater number of border rows to comply with the 0.9% threshold, and an updated extension to the model is provided to quantify these effects.

Dominance in the Tetra Pak case: an empirical approach

The 1991 decision of the European Commission on the Tetra Pak case was based on information which seemed to prove the firm's anti-competitive behavior. The Tetra Pak case is investigated here focusing on the meaning of multimarket dominance, using empirical techniques. We find that a more rigorous analysis of the data available would not confirm the Commission's assertions. That is, it cannot be concluded with certainty that the Commission was right to relate Tetra Pak's dominance in the aseptic sector to its market power in the non-aseptic sector. Our results suggest a general framework for the analysis of abusive transfer of market power across vertically or/and horizontally related markets.

A simple syllogism-solving test: empirical findings and implications for g research

It has been reported that the ability to solve syllogisms is highly g-loaded. In the present study, using a self-administered shortened version of a syllogism-solving test, the BAROCO Short, we examined whether robust findings generated by previous research regarding IQ scores were also applicable to BAROCO Short scores. Five syllogism-solving problems were included in a questionnaire as part of a postal survey conducted by the Keio Twin Research Center. Data were collected from 487 pairs of twins (1021 individuals) who were Japanese junior high or high school students (ages 13–18) and from 536 mothers and 431 fathers. Four findings related to IQ were replicated: 1) The mean level increased gradually during adolescence, stayed unchanged from the 30s to the early 50s, and subsequently declined after the late 50s. 2) The scores for both children and parents were predicted by the socioeconomic status of the family. 3) The genetic effect increased, although the shared environmental effect decreased during progression from adolescence to adulthood. 4) Children's scores were genetically correlated with school achievement. These findings further substantiate the close association between syllogistic reasoning ability and g.

Responding to the EU Emissions Trading Scheme: an empirical analysis of corporate carbon disclosure strategy

The purpose of this paper is to explore how companies that hold carbon trading accounts under European Union Emissions Trading Scheme (EU ETS) respond to the climate change by using disclosures on carbon emissions as a means to generate legitimacy compared to others. The study is based on disclosures made in annual reports and stand-alone sustainability reports of UK listed companies from 2001- 2012. The study uses content analysis to capture both the quality and volume of the carbon disclosures. The results show that there is a significant increase in both the quality and volume of the carbon disclosures after the launch of EU ETS. Companies with carbon trading accounts provide greater detailed disclosures as compared to the others without an account. We also find that company size is positively correlated with the disclosures while the association with the industry produces an inconclusive result.