Bayesian averaging of classical estimates in asymmetric vector autoregressive models

The estimated vector autoregressive (VAR) model is sensitive to model misspecifications, resulting to biased and inconsistent parameter estimates. This article extends the Bayesian averaging of classical estimates, a robustness procedure in cross-section data, to a vector time-series that is estimated using a large number of asymmetric VAR models. The proposed procedure was applied to simulated data from various forms of model misspecifications. The results of the simulation suggest that, under misspecification problems, particularly if an important variable and moving average (MA) terms were omitted, the proposed procedure gives robust results and better forecasts than the automatically selected equal lag-length VAR model.     

Social-ecological innovation: adaptive responses to urban environmental conditions

Novel approaches to natural resource management, particularly those which promote stakeholder participation, have been put forward as fundamental ingredients for establishing resilient, polycentric forms of environmental governance. This is nowhere more pertinent than in the case of the complex adaptive systems associated with urban areas. Decentralisation of urban green space management has been posited as an element thereof which, according to resilience thinking, should contribute to the adaptive capacity of cities and the ecosystem services upon which they rely. Implicit in this move towards increased adaptive capacity is the ability to manage through innovation. Although the importance of innovation towards system adaptability has been acknowledged, little work has thus far been carried out which demonstrates that innovative use of urban green space represents a form of adaptive response to environmental conditions. The current paper reports on research which maps examples of organised social-ecological innovation (OSEI) in an urban study area and evaluates them as adaptive responses to local environmental conditions which may contribute to system resilience. The results present OSEI as a coherent body of responses to local social and environmental deprivation, exhibiting diversity and adaptability according to individual contexts. The study therefore provides evidence for the importance of local stakeholder-led innovation as in the building of adaptive capacity in urban social-ecological systems.     

Alternative tests for heteroscedasticity of disturbances:a comparative study

This paper presents three small sample tests for testing the heteroscedasticity among regression disturbances. The power of these tests are compared with two of the leading tests for this hypothesis, one by Goldfeld and Quandt [5] and the other by Theil [17]. We also provide a heuristic method of selecting the number of middle observations to be deleted for the Goldfeld-Quandt type of tests.     

Learning from a lot: Empirical Bayes for high‐dimensional model‐based prediction

Empirical Bayes is a versatile approach to “learn from a lot” in two ways: first, from a large number of variables and, second, from a potentially large amount of prior information, for example, stored in public repositories. We review applications of a variety of empirical Bayes methods to several well‐known model‐based prediction methods, including penalized regression, linear discriminant analysis, and Bayesian models with sparse or dense priors. We discuss “formal” empirical Bayes methods that maximize the marginal likelihood but also more informal approaches based on other data summaries. We contrast empirical Bayes to cross‐validation and full Bayes and discuss hybrid approaches. To study the relation between the quality of an empirical Bayes estimator and p, the number of variables, we consider a simple empirical Bayes estimator in a linear model setting. We argue that empirical Bayes is particularly useful when the prior contains multiple parameters, which model a priori information on variables termed “co‐data”. In particular, we present two novel examples that allow for co‐data: first, a Bayesian spike‐and‐slab setting that facilitates inclusion of multiple co‐data sources and types and, second, a hybrid empirical Bayes–full Bayes ridge regression approach for estimation of the posterior predictive interval.     

The influence of ‘topic and resource’ on some aspects of social theorising

Developments in sociological theory since the 1960s have been responses to disciplinary problems rather than changes in fashion. The problem of topic and resource—where sociology has to use everyday understandings and practices as study resources even though they are legitimate topics of enquiry—has been an important and sometimes neglected spur to many of these developments. The turn to discourse, conversation analysis and the rise of Bourdieu's reflexivity are all attempts to address the problem, but each is shown to be unsatisfactory in different ways. In summary, they seek to address the issue as requiring either a principled methodological or a principled theoretical solution, and neither approach is capable of comprehensively addressing the matter. It is argued that these ‘solutions’ depend, in turn, on one of two particular construals of what the ‘problem’ consists in, neither of which is necessary or coherent. Each, it is argued, depends on a philosophical trick: making language out to need formal improvement (the Bertrand Russell trick) or introducing inappropriate scepticism to everyday life (the René Descartes trick). It is suggested that treating topic and resource not as a problem but as something which opens up new areas of investigation successfully deflates the issue and avoids unnecessary theoretical and methodological contortions.