A play-the-winner-type urn design with reduced variability

We propose a new adaptive allocation rule, the drop-the-loser, that randomizes subjects in the course of a trial comparing treatments with dichotomous outcomes. The rule tends to assign more patients to better treatments with the same limiting proportion as the randomized play-the-winner rule. The new design has significantly less variable allocation proportion than the randomized play-the-winner rule. Decrease in variability translates into a gain in statistical power. For some values of success probabilities the drop-the-loser rule has a double advantage over conventional equal allocation in that it has better power and assigns more subjects to the better treatment. Acknowledgments. I thank Stephen Durham, the associate editor, and the referees for their helpful suggestions.     

Performance assessment of ensemble learning systems in financial data classification

Financial data classification plays an important role in investment and banking industry with the purpose to control default risk, improve cash and select the best customers. Ensemble learning and classification systems are becoming gradually more applied to classify financial data where outputs from different classification systems are combined. The objective of this research is to assess the relative performance of existing state‐of‐the‐art ensemble learning and classification systems with applications to corporate bankruptcy prediction and credit scoring. The considered ensemble systems include AdaBoost, LogitBoost, RUSBoost, subspace, and bagging ensemble system. The experimental results from three datasets: one is composed of quantitative attributes, one encompasses qualitative data, and another one combines both quantitative and qualitative attributes. By using ten‐fold cross‐validation method, the experimental results show that AdaBoost is effective in terms of low classification error, limited complexity, and short time processing of the data. In addition, the experimental results show that ensemble classification systems outperform existing models that were recently validated on the same databases. Therefore, ensemble classification system can be employed to increase the reliability and consistency of financial data classification task.     

Remittances of Immigrants in Greece

The present paper is a case study of the remittance behavior of immigrants in Greece. In addition to some of the usual factors affecting remittance behavior, it examines five factors than have received little attention in the literature: relative deprivation, existence of property in the country of origin, expectation on inheritance, stability of employment, and more than one migrant in the family. We use Heckman's selection model with survey data from 607 immigrants. Significant factors for the decision to remit are the existence of children living with the migrant, stability of employment, and gender. The basic factors affecting the size of remittances are the level of income, the existence of children living with the migrant, and the length of time in Greece.     

Estimation of dynamic panel data models with sample selection

We propose a new method for estimating dynamic panel data models with selection. The method uses backward substitution for the lagged dependent variable, which leads to an estimating equation that requires correcting for contemporaneous selection only. The estimator is valid under relatively weak assumptions about errors and permits avoiding the weak instruments problem associated with differencing. We also propose a simple test for selection bias that is based on the addition of a selection term to the first‐difference equation and subsequent testing for significance of this term. The methods are applied to estimating dynamic earnings equations for women. Copyright © 2011 John Wiley & Sons, Ltd.     

Imputing relevant information from multi-day GPS tracers for retail planning and management using data fusion and context-sensitive learning

It is well known that the right location of shopping centres is of paramount importance. Unless stores succeed in attracting their own clientele, they rely to a large extent on the impulse behaviour of shoppers. To evaluate alternative locations, models of pedestrian behaviour may be useful. Modern technologies such as GPS and RFID offer new possibilities providing data on routes and stops, which are required as input for such models. An automatic interpretation of GPS tracers with respect to the activities being conducted could enhance the applicability of such technologies to retail management applications. This paper reviews this rapidly growing literature, and shows how automatic data imputation can be established by using Bayesian belief networks and how GPS traces can be fused with land use data of retail location.     

Estimating panel data models in the presence of endogeneity and selection

We consider estimation of panel data models with sample selection when the equation of interest contains endogenous explanatory variables as well as unobserved heterogeneity. Assuming that appropriate instruments are available, we propose several tests for selection bias and two estimation procedures that correct for selection in the presence of endogenous regressors. The tests are based on the fixed effects two-stage least squares estimator, thereby permitting arbitrary correlation between unobserved heterogeneity and explanatory variables. The first correction procedure is parametric and is valid under the assumption that the errors in the selection equation are normally distributed. The second procedure estimates the model parameters semiparametrically using series estimators. In the proposed testing and correction procedures, the error terms may be heterogeneously distributed and serially dependent in both selection and primary equations. Because these methods allow for a rather flexible structure of the error variance and do not impose any nonstandard assumptions on the conditional distributions of explanatory variables, they provide a useful alternative to the existing approaches presented in the literature.     

A Crisis of the Overcrowded Future: Shadow Banking and the Political Economy of Financial Innovation

This article focuses on the role the shadow banking system played in the financial crisis of 2007–9. Engaging with emergent theories of shadow banking, I inquire into its structural role in contemporary capitalism. My main premise here is that the crisis of 2007–9 is distinct in financial history because it did not centre on any organised market. Rather, it was crisis of the overcrowded financial channels bridging the present and the future, which have become congested because of the massive concentration of financial values generated, yet not sustained, through the shadow banking network. My analysis suggests that shadow banking has determined the nature of financial crisis of 2007–9 and continues to play a necessary role in financial capitalism based on futurity. Drawing on scholarship in financial Keynesianism, contemporary legal studies and early evolutionary political economy, I argue that shadow banking is best seen as the organic institutional infrastructure of financialised capitalism based on debt and geared towards futurity, a concept originally developed by John Commons.     

Does a Nearby Murder Affect Housing Prices and Rents? The Case of Sydney

We measure the impact of murders on prices and rents of homes in Sydney. We find that housing prices fall by 3.9 per cent for homes within 0.2 miles of the murder in the year following the murder, and weaker results in the second year after a murder. We do not find any effects of murders on rents. Higher media coverage and being located closer to the murder (within 0.1 mile) have no additional effect on prices. Taken together, our findings suggest that proximity to a murder affects nearby property prices, particularly in the first year after the incident.     

State intervention and the microcredit market: the role of business development services

We analyze in this paper how various forms of state intervention can impact microfinance institutions’ lending behavior. Using a simple model where entrepreneurs receive individual uncollateralized loans, we show that, not surprisingly, state intervention through the loan guarantee increases the number of entrepreneurs receiving a loan. However, after modeling business development services (BDS) provided by the microfinance institution, we show that the loan guarantee can have a counterproductive effect by reducing the number of entrepreneurs benefiting from such services. We therefore analyze an alternative policy: BDS subsidization. We show that if BDS are efficient enough and are targeted toward less performing borrowers, then—for fixed government expenditures—such subsidies do better in terms of financial inclusion than the loan guarantee. Moreover, we argue that—under similar conditions—BDS subsidization alone does better in terms of financial inclusion than a mix of policies.     

Focus groups as a tool in the construction of questionnaires: the case of discriminatory attitudes

The aim of the present article is twofold First, it tests and demonstrates the supplementary use of focus groups to construct quantitatively oriented survey on anti-minority sentiment. Second, it clarifies two major theoretical concepts-prejudice versus perceived threat-in the research on discriminatory attitudes towards minority populations. More specifically, by using data from focus group discussions on foreign workers in Israel, the study examines the use of such groups for measurement of prejudice and perceived threat within the social and cultural context of Israeli society. The analysis of focus group discussions provides adapted questionnaire scales for measuring prejudice and perceived threat. The findings support the premise that prejudice and perceived threat are two distinct theoretical concepts.