Combining least-squares and quantile regressions

Least-squares and quantile regressions are method of moments techniques that are typically used in isolation. A leading example where efficiency may be gained by combining least-squares and quantile regressions is one where some information on the error quantiles is available but the error distribution cannot be fully specified. This estimation problem may be cast in terms of solving an over-determined estimating equation (EE) system for which the generalized method of moments (GMM) and empirical likelihood (EL) are approaches of recognized importance. The major difficulty with implementing these techniques here is that the EEs associated with the quantiles are non-differentiable. In this paper, we develop a kernel-based smoothing technique for non-smooth EEs, and derive the asymptotic properties of the GMM and maximum smoothed EL (MSEL) estimators based on the smoothed EEs. Via a simulation study, we investigate the finite sample properties of the GMM and MSEL estimators that combine least-squares and quantile moment relationships. Applications to real datasets are also considered.     

Making the Technological Transition – Citizens' Encounters with Digital Pension Services

We examine in this article the impact of digitization on the provision of public services by studying how citizens experience the use of web portals in their pension planning. Based on focus groups and user test material, we elucidate five critical phases that each operate as an obstacle for citizens' further engagement in the digital self‐service process: interest, access, comprehension, reflection and support. We argue that these phases and the obstacles they entail illustrate a transition away from a situation in which control and agency over the quality of public services is embedded in a dyadic relationship between citizens and frontline personnel, and to a situation characterized by more complex relations between citizens, the front line, and the digital infrastructure. We argue that this transition implies that citizens are required to possess a new type of competence that contains both financial and digital skills. Citizens who are unable to develop or acquire such competence are likely to be disadvantaged by the services. Lastly, we argue that these developments pose significant challenges for public administrations to ensure the overall quality of the public services.     

Addressing common forms of child maltreatment: evidence‐informed interventions and gaps in current knowledge

This paper reviews interventions for preventing the occurrence and recurrence of major types of child maltreatment. We begin with an overview of the challenges of establishing evidence‐based interventions to prevent child abuse and neglect in many countries, and underscore the importance of this need with child maltreatment incidence rates in the USA, and how much each type and subtype contribute to child out‐of‐home placement. Next, we identify the well‐supported, supported and promising interventions for each child maltreatment type and subtype, according to their level of research evidence using an evidence‐based clearing house. The paper closes with a discussion of the implications for practice, evaluation, policy and agency management, including intervention knowledge gaps that showcase areas that need additional practice research.     

A Smooth Bootstrap Procedure towards Deriving Confidence Intervals for the Relative Risk

Given a pair of sample estimators of two independent proportions, bootstrap methods are a common strategy towards deriving the associated confidence interval for the relative risk. We develop a new smooth bootstrap procedure, which generates pseudo-samples from a continuous quantile function. Under a variety of settings, our simulation studies show that our method possesses a better or equal performance in comparison with asymptotic theory based and existing bootstrap methods, particularly for heavily unbalanced data in terms of coverage probability and power. We illustrate our procedure as applied to several published data sets.     

Market-Based Credit Ratings

We present a methodology for rating in real-time the creditworthiness of public companies in the U.S. from the prices of traded assets. Our approach uses asset pricing data to impute a term structure of risk neutral survival functions or default probabilities. Firms are then clustered into ratings categories based on their survival functions using a functional clustering algorithm. This allows all public firms whose assets are traded to be directly rated by market participants. For firms whose assets are not traded, we show how they can be indirectly rated by matching them to firms that are traded based on observable characteristics. We also show how the resulting ratings can be used to construct loss distributions for portfolios of bonds. Finally, we compare our ratings to Standard & Poors and find that, over the period 2005 to 2011, our ratings lead theirs for firms that ultimately default.     

Geographically Weighted Negative Binomial Regression—incorporating overdispersion

Global regression assumes that a single model adequately describes all parts of a study region. However, the heterogeneity in the data may be sufficiently strong that relationships between variables can not be spatially constant. In addition, the factors involved are often sufficiently complex that it is difficult to identify them in the form of explanatory variables. As a result Geographically Weighted Regression (GWR) was introduced as a tool for the modeling of non-stationary spatial data. Using kernel functions, the GWR methodology allows the model parameters to vary spatially and produces non-parametric surfaces of their estimates. To model count data with overdispersion, it is more appropriate to use a negative binomial distribution instead of a Poisson distribution. Therefore, we propose the Geographically Weighted Negative Binomial Regression (GWNBR) method for the modeling of data with overdispersion. The results obtained using simulated and real data show the superiority of this method for the modeling of non-stationary count data with overdispersion compared with competing models, such as global regressions, e.g., Poisson and negative binomial and Geographically Weighted Poisson Regression (GWPR). Moreover, we illustrate that these competing models are special cases of the more robust model GWNBR.     

Analysis and Implications of Retraction Period and Coauthorship of Fraudulent Publications

Studies have indicated that the number and frequency of fraudulent publications being retracted are not subsiding even with greater awareness of such incidents in the recent decades. In this study, the trends of retraction period, number of citations and coauthors of 5 selected researchers who had ≥ 15 fraudulent publications retracted were analyzed. The obtained results shows that the retraction period is 48.96 ± 32.16 months for the 113 publications affiliated to the 5 studied researchers. There are a total of 180 coauthors with 6.40 ± 3.26 coauthors per researcher's retracted publication. The linear regression analysis indicates that there is limited correlation (R² = .008) between the citation counts and retraction period. The p value for multiple F-tests to assess the number of coauthors to a fraudulent publication on an interresearcher basis is found to be ranging from < .001 to .458. It is also found that a better correlation (R² = .592) exists between the likelihood of a researcher to involve different individuals for isolated fraudulent publications while only selecting very few to be their frequent coauthors of their mischievous acts. With this study, the possible use of coauthors as a strategy for publishing fraudulent work and a potential approach to tighten coauthorship are discussed.     

Notes on two sample tests for partially correlated (paired) data

We provide several methods to compare two Gaussian distributed means in the two sample location problems under the assumption of partially dependent observations. Simulation studies indicate that our test procedure is frequently more powerful than other methods depending on the ratio of the unpaired data and the strength and direction of the correlation between the two variables. The tests used in our comparative study are illustrated with an example based on data from a small gynecological study.     

Model‐based Prediction In Ecological Surveys Including Those with Incomplete Detection

This paper explores and develops model‐based predictors for surveys of plants and wildlife including those with incomplete detection. The methodology allows for estimating a detection function to account for objects which were not detected at the time of the survey. The model‐based theory utilises generalized linear models (GLMs) and is either new or adapted from other areas of sampling. A simulation study is used to validate the estimators and comparisons are made with an integrated likelihood approach. An aerial survey of kangaroos in western New South Wales is used to illustrate the theory. The area within 50m of the aircraft is treated as a strip transect and mark‐recapture methods are used to estimate the detection function.