Robust Small Area Estimation under Spatial Non‐stationarity

The effective use of spatial information in a regression‐based approach to small area estimation is an important practical issue. One approach to account for geographic information is by extending the linear mixed model to allow for spatially correlated random area effects. An alternative is to include the spatial information by a non‐parametric mixed models. Another option is geographic weighted regression where the model coefficients vary spatially across the geography of interest. Although these approaches are useful for estimating small area means efficiently under strict parametric assumptions, they can be sensitive to outliers. In this paper, we propose robust extensions of the geographically weighted empirical best linear unbiased predictor. In particular, we introduce robust projective and predictive estimators under spatial non‐stationarity. Mean squared error estimation is performed by two analytic approaches that account for the spatial structure in the data. Model‐based simulations show that the methodology proposed often leads to more efficient estimators. Furthermore, the analytic mean squared error estimators introduced have appealing properties in terms of stability and bias. Finally, we demonstrate in the application that the new methodology is a good choice for producing estimates for average rent prices of apartments in urban planning areas in Berlin.     

Stuck at a workplace: What’s work control,demands and learning got to do with it? A longitudinal multilevel study on Swedish permanent employees in situations of ‘workplace locked-in’

Abstract

Whilst health consequences of being locked-in at the workplace have been documented in several research studies, it is largely unknown how work characteristics and their changes over time relate to risks for becoming locked-in at a certain workplace. Accordingly, this paper studied how perceived control, learning opportunities and quantitative demands at work associate with workplace-locked-in (WPLI). The study included permanent employees who participated in the Swedish Longitudinal Occupational Survey of Health (SLOSH) study in wave 3 through 5 (n = 2918 individuals; n = 7460 observations). Results from multi-level analysis show that there was significant individual variation in WPLI changes over time, even though on average, WPLI decreased slightly. Differences in work characteristics between individuals (L2) and across time (L1) associated significantly with WPLI: higher levels of job control and learning opportunities related to lower odds ratios for WPLI, whereas higher quantitative job demands associated with higher odds ratios of WPLI. Moreover, differences in quantitative job demands, number of job changes and educational achievements explained the individual variations of WPLI developments over time. The result shows that WPLI can – to some extent – be prevented or reduced through good work design, and implications for HR managers and organizations are discussed.     

Existence of Edgeworth equilibria for economies with asymmetric information

We analyze a general equilibrium model with asymmetrically informed agents. A basic and well-known consequence of asymmetric information is the failure of equal treatment property. Despite of this, we show that in every replica economy a private core allocation, which treats agents of the same type equally, exists. This result is provided in the framework of two different models, with and without expectations, and it suffices to guarantee in both cases the existence of a private Edgeworth equilibrium. The author is grateful to an anonymous referee for his/her careful reading and helpful comments and acknowledges support by Research Grant no. 2006132713-005 (Ministero Italiano dell’Università e della Ricerca).     

Deterrence, Lawsuits, and Litigation Outcomes Under Court Errors

This article presents a strategic model of liability and litigationunder court errors. Our framework allows for endogenous choiceof level of care and endogenous likelihood of filing and disputes.We derive sufficient conditions for a unique universally divinemixed-strategy perfect Bayesian equilibrium under low courterrors. In this equilibrium, some defendants choose to be grosslynegligent; some cases are filed; and some lawsuits are dropped,some are resolved out of court, and some go to trial. We findthat court errors in the size of the award, as well as damagecaps and split awards, reduce the likelihood of trial but increasefiling and reduce the deterrence effect of punitive damages.We derive conditions under which the adoption of the Englishrule for allocating legal costs reduces filing.     

Information and size of coalitions

We consider a set of asymmetrically informed agents, where the information of each trader is susceptible of being altered when she becomes a member of a coalition. For this, we consider a general rule that depending on the coalition, a signal (or an information partition) is assigned to each member of the coalition. We set examples showing that Grodal’s (Econometrica 40:581–583, 1972), Schmeidler’s (Econometrica 40:579–580, 1972) and Vind’s (Econometrica 40:585–586, 1972) core characterizations of a continuum economy may fail in this general informational setting. However, under mild assumptions on the rule, we extend Schmeidler’s and Vind’s results to economies that allocate information to agents in each coalition according to the rule. We then focus on information mechanisms based on the size of coalitions and provide a general characterization result for the corresponding cores. Moreover, we pay close attention to the rule that assigns the shared information to each member of specific coalitions. We prove that the resulting cores are exactly the same independently of whether arbitrarily small or large coalitions share information.     

Wild horse-based tourism as wildlife tourism: the wild horse as the other

Wild horses as the focus of tourism occupy a unique position. This is a consequence of their ambiguous status in the natural and cultural landscape, particularly in North America. Wild horses are ecological agents, cultural icons, economic factors and political pawns. The complexity of their management environment has an impact on the tourism and recreational context. Focusing on the western US and western Canada, this article explores a conceptual framework for wild horse-based tourism and highlights unique characteristics of the encounter between wild horses and visitors, drawing on literature and empirical data. It positions wild horse-based tourism within a framework of wildlife tourism and introduces the wild horse as a charismatic animal which elicits strong reactions from visitors who encounter it. While visitors tend to embrace the wild horse as an integral part of its habitat, as a symbol of the western frontier, and an embodiment of freedom, the animal remains an extremely polarising subject in the management debate of public lands in the USA and Canada. The wild horse's beleaguered status in both countries seriously interferes with the realisation of the true potential of wild horse-based tourism. On the other hand, wild horse supporters pin high hopes on this industry's transformative power.     

Scope and scale economies in merging hospitals prior to merger

During the 1980s, the impact of the Prospective Payment System and increased pressure from payers for discounts placed pressure on hospitals to become more efficient. An increase in the number of hospital mergers during this period suggests that merger was a strategy used by hospitals to improve efficiency. This hypothesis is tested by estimating multiproduct scope and scale economies in merging hospitals one year prior to the merger and comparing these with a group of controls using estimates from a hybrid translog cost function. Diseconomies of scale were present when the estimated hybrid translog cost function is evaluated at variable means. The results suggest that merging hospitals could obtain operational efficiencies through merger.