Estimation, Learning and Parameters of Interest in a Multiple Outcome Selection Model

We describe estimation, learning, and prediction in a treatment-response model with two outcomes. The introduction of potential outcomes in this model introduces four cross-regime correlation parameters that are not contained in the likelihood for the observed data and thus are not identified. Despite this inescapable identification problem, we build upon the results of Koop and Poirier (1997) to describe how learning takes place about the four nonidentified correlations through the imposed positive definiteness of the covariance matrix. We then derive bivariate distributions associated with commonly estimated “treatment parameters” (including the Average Treatment Effect and effect of Treatment on the Treated), and use the learning that takes place about the nonidentified correlations to calculate these densities. We illustrate our points in several generated data experiments and apply our methods to estimate the joint impact of child labor on achievement scores in language and mathematics.     

Improving routing and scheduling decisions at a distributor of industrial gasses

This paper investigates cyclical inventory replenishment for a company's regional distribution center that supplies, distributes, and manages inventory of carbon dioxide (_CO2)$({\mathrm{CO}}_2)$ at over 900 separate customer sites in Indiana. The company previously experienced high labor costs with excessive overtime and maintained a regular back-log of customers experiencing stockouts. To address these issues we implemented a three-phase heuristic for the cyclical inventory routing problem encountered at one of the company's distribution centers. This heuristic determines regular routes for each of three available delivery vehicles over a 12-day delivery horizon while improving four primary performance measures: delivery labor cost, stockouts, delivery regularity, and driver–customer familiarity. It does so by first determining three sets of cities (one for each delivery vehicle) that must be delivered to each day based on customer requirements. Second, the heuristic assigns the remaining customers in other cities to one of the three “backbone routes” determined in phase 1. And third, it balances customer deliveries on each daily route over the schedule horizon. Through our methodology, we were able to significantly reduce overtime, driving time, and labor costs while improving customer service.     

Liability of foreignness as a barrier to knowledge spillovers: Lost in translation?

Multinational corporations (MNC) search increasingly for lead market knowledge and technological expertise around the globe. We investigate whether their subsidiaries gain access to these valuable sources of host country knowledge to the same degree as domestic rivals. We develop a theoretical framework for “why” and “how” a lack of embeddedness and legitimacy (liability of foreignness) may translate into additional obstacles for foreign subsidiaries. We test these hypotheses empirically using a broad dataset of more than 1100 firms in Germany. We find that MNCs can compete on an equal footing with host country competitors when it comes to generating impulses for innovations from universities. They are significantly challenged by liabilities of foreignness, though, when host country customers are involved. The disadvantages are especially pronounced when the host country industry is at the technological forefront. We suggest that the disadvantages arising from liability of foreignness in the host country are especially relevant when promising lead customers have to be identified and their tacit and often unarticulated impulses have to be transferred, understood and prioritized. Management recommendations are developed on the basis of these results.     

Reversible jump, birth-and-death and more general continuous time Markov chain Monte Carlo samplers

Summary. Reversible jump methods are the most commonly used Markov chain Monte Carlo tool for exploring variable dimension statistical models. Recently, however, an alternative approach based on birth-and-death processes has been proposed by Stephens for mixtures of distributions. We show that the birth-and-death setting can be generalized to include other types of continuous time jumps like split-and-combine moves in the spirit of Richardson and Green. We illustrate these extensions both for mixtures of distributions and for hidden Markov models. We demonstrate the strong similarity of reversible jump and continuous time methodologies by showing that, on appropriate rescaling of time, the reversible jump chain converges to a limiting continuous time birth-and-death process. A numerical comparison in the setting of mixtures of distributions highlights this similarity.     

Contested control at the margins of the state

ABSTRACT

While the fact that the implementation of migration policies fails to perfectly manage migration is well known, the actual dynamics of policy implementation have received little attention to date. A serious engagement with this phenomenon requires a move beyond policy texts and political intentions, and towards a ‘migration regime’ perspective that pays attention to the inherent contradictions, conflicts of interest and competing logics within migration control practices. This collection posits a multi-actor perspective that includes state agents, migrants and non-state actors alike and proposes three key factors that require a closer examination: competing institutional logics, discretionary practices and migrants’ agency. Based on original empirical research, the contributions of this collection ‘zoom in’ on specific asymmetrical negotiations over the right to enter or remain in Europe, and focus on the institutional logics and interplay between the different actors involved.     

Refusal Bias in the Estimation of HIV Prevalence

In 2007, UNAIDS corrected estimates of global HIV prevalence downward from 40 million to 33 million based on a methodological shift from sentinel surveillance to population-based surveys. Since then, population-based surveys are considered the gold standard for estimating HIV prevalence. However, prevalence rates based on representative surveys may be biased because of nonresponse. This article investigates one potential source of nonresponse bias: refusal to participate in the HIV test. We use the identity of randomly assigned interviewers to identify the participation effect and estimate HIV prevalence rates corrected for unobservable characteristics with a Heckman selection model. The analysis is based on a survey of 1,992 individuals in urban Namibia, which included an HIV test. We find that the bias resulting from refusal is not significant for the overall sample. However, a detailed analysis using kernel density estimates shows that the bias is substantial for the younger and the poorer population. Nonparticipants in these subsamples are estimated to be three times more likely to be HIV-positive than participants. The difference is particularly pronounced for women. Prevalence rates that ignore this selection effect may be seriously biased for specific target groups, leading to misallocation of resources for prevention and treatment.