Cross-validating fit and predictive accuracy of nonlinear quantile regressions

The paper proposes a cross-validation method to address the question of specification search in a multiple nonlinear quantile regression framework. Linear parametric, spline-based partially linear and kernel-based fully nonparametric specifications are contrasted as competitors using cross-validated weighted L ₁-norm based goodness-of-fit and prediction error criteria. The aim is to provide a fair comparison with respect to estimation accuracy and/or predictive ability for different semi- and nonparametric specification paradigms. This is challenging as the model dimension cannot be estimated for all competitors and the meta-parameters such as kernel bandwidths, spline knot numbers and polynomial degrees are difficult to compare. General issues of specification comparability and automated data-driven meta-parameter selection are discussed. The proposed method further allows us to assess the balance between fit and model complexity. An extensive Monte Carlo study and an application to a well-known data set provide empirical illustration of the method.     

On Empirical Processes for Ergodic Diffusions and Rates of Convergence of M-estimators

Abstract. This paper contributes to the development of empirical process theory for ergodic diffusions. We prove an entropy‐type maximal inequality for the increments of the empirical process of an ergodic diffusion. The inequality is used to study the rate of convergence of M‐estimators.     

Homesteading in the Wild West: An Appreciation of the Pacific Sociological Association

In this very personal essay, I suggest four aspects of the PSA meetings which contributed to the creation of a “wild west” atmosphere, namely the methods used for the creation of sessions, the content of the sessions, the high level of social interaction, and the focus on student involvement. I conclude with a concern about the increased focus on making the meetings more prestigious and professional.     

Making urban sector lending work; lessons from a capacity building programme in Karnataka, India

This article presents experiences with human resource development and ‘capacity building programmes’ in urban sector lending by multilateral agencies. The article first reviews the history of ‘capacity building’ and provides a view on its meaning. The experiences with a capacity building programme, part of an infrastructure loan to the state of Karnataka in India, are subsequently elaborated as an illustration. The article seeks to identify the factors determining the success of capacity building programmes which are part of urban lending. It concludes first that Terms of Reference, drafted in a bottom–up and demand-driven fashion are a better guarantee for ownership, commitment and positive results. Secondly, capacity building activities need to be tailored to the ability of the towns to make changes in their operation. Finally, capacity building at local government level can only be effective if reciprocated by supportive measures of state governments. The paper also argues that Technical Assistance (TA) as a component of capacity building should be more strongly integrated in the loan component activities and continue throughout the project period.     

Scholarly Productivity of Social Work Faculty at Historically Black Colleges and Universities: Are h-Index Scores a Suitable Measure?

Faculty scholarship at historically Black colleges and universities (HBCU) has in the past served as a blueprint for the Black masses. Even today, HBCU faculty scholarship continues to be an informative source to communicate accurate information regarding marginalized groups. This study examines h-index scores of 65 faculty members at five doctorate-granting schools of social work. The majority of calculated h-index scores were considered to be low in terms of productivity. We make the argument that these scores are not a good measure of productivity because of the problematic nature of their use to evaluate HBCU faculty. Implications for future research, practice, and teaching are presented.