The FastHCS algorithm for robust PCA

Principal component analysis (PCA) is widely used to analyze high-dimensional data, but it is very sensitive to outliers. Robust PCA methods seek fits that are unaffected by the outliers and can therefore be trusted to reveal them. FastHCS (high-dimensional congruent subsets) is a robust PCA algorithm suitable for high-dimensional applications, including cases where the number of variables exceeds the number of observations. After detailing the FastHCS algorithm, we carry out an extensive simulation study and three real data applications, the results of which show that FastHCS is systematically more robust to outliers than state-of-the-art methods.     

Effort-reward imbalance and work-home interference: a two-wave study among European male nurses

ABSTRACT

This one-year follow-up study among 1,421 male nurses from seven European countries tested the validity of the Effort-Reward Imbalance (ERI) model in predicting prospective vital exhaustion and work-home interference. We hypothesised that effort and lack of reward would have both main and interactive effects on future outcomes. Results of structural equation modelling (SEM) showed that effort was positively related to exhaustion and work-home interference, both simultaneously and over time. Lack of reward predicted increased exhaustion at follow-up, but effort-reward imbalance did not influence the outcomes. Additionally, Time 1 exhaustion predicted increased work-home interference and exhaustion at follow-up. These results do not support the ERI model, which postulates a primacy of effort-reward imbalance over main effects. Instead, the findings are in line with dual path models of job stress and work-home interference. Multi-group SEM showed partial cross-cultural metric invariance for the ERI measure of effort, but the ERI measure of rewards showed no metric measurement invariance, indicating its meaning is qualitatively different across cultures. Nevertheless, the main conclusions were markedly similar for each national sub-sample. We discuss the theoretical and practical implications of our study.     

Exploring teachers’ career self-management. Considering the roles of organizational career management,occupational self-efficacy,and learning goal orientation.

The current study examines how organizational career management – i.e. activities undertaken by schools in order to plan and manage teachers’ careers – relates to teachers’ career self-management – i.e. teachers steering their careers by means of searching for opportunities, networking, or seeking supervisory support. Moreover, it examines the mediating roles of occupational self-efficacy and learning goal orientation in this relationship. Mediation analysis in SPSS, using the PROCESS macro of survey data from 220 Dutch secondary school teachers, showed that positive relationships between organizational career management and career self-management were mediated by occupational self-efficacy and learning goal orientation.     

The Relative Fit measure for evaluating a blockmodel

Statistical Methods & Applications - A blockmodel is a network in which the nodes are clusters of equivalent (in terms of the structure of the links connecting) nodes in the network being...     

Will the Dismantlement of Early Retirement Schemes Increase Older Unemployment? A Competing‐risk Analysis for Belgium

On the basis of administrative data for Belgium, we estimate a competing‐risk model on transitions from employees aged 50 and older into unemployment, early and old‐age retirement while accounting for forward‐looking work disincentives. Our estimates are used to simulate a cut in early retirement benefits. Although this would enhance the financial sustainability of the social security system, our simulations predict a strong increase in unemployment among older blue‐collar workers in traditional industries. Members of private saving plans or occupational pension schemes and highly educated workers are predicted to move into the old‐age pension system.     

Perceived Quality of Life and Self-Rated Health among First-Year University Students

This study compares health status and qualityof life assessments of first-year universitystudents with those of their same-age workingcounterparts. Subjects and materials for eachgroup were gathered in 1999 from twocross-sectional data sets from the Swedishregion of Östergötland, covering malesand females aged 20–34 years. Subjects'perceived quality of life (QoL) and self-ratedhealth (SRH) were assessed on a 10-point scale(Ladder scale) and a five-point scale,respectively. Gender-based comparison revealedthat, for both males and females, first-yearuniversity students' average perceived QoL waslower than that of their working counterparts(p < 0.0001 in all instances). A higherproportion of students than expected ratedtheir health as ``average' or as ``low'(p < 0.0001). Perceived QoL was significantlycorrelated with SRH in both groups. Differencesin perceived QoL and SRH exist between studentsand their full-time working peers, and thedeterminants of these differences deservesgreater attention. Knowledge of thedeterminants of SRH and perceived QoL amonguniversity students might then be translatedinto sound and effective public-health practiceand intervention programs.     

Poverty persistence among the elderly in the transition from work to retirement

On the basis of matched Income Tax/Census data, the majority of the elderly poor in Belgium appear to be persistently poor. The question arises as to why this might be so. To the extent that individual characteristics such as abilities or tastes persist over time, these may also be the reason that individuals persist in poverty over time. In that case, one expects that once individual characteristics are controlled for, duration dependence in poverty becomes spurious. The alternative possibility is that the poverty experience itself has a causal impact on future poverty. This may be because of work disincentives, lack of investment in human capital and/or loss of motivation during the period of job search. The reasons for dependence in poverty are of interest for developing effective poverty-reducing policies. The estimation of a multiple-spell hazard model of transitions in and out of poverty, that allows for unobserved heterogeneity and a significant initial condition problem, supports the suggestion of true duration dependence in poverty. In Belgium the elderly unemployed are exempted from job search and lack any training or counselling from government agencies, while workers above 50 are given strong financial incentives to leave the labour market as soon as possible through different kinds of early retirement schemes.     

New analysis and computational study for the planar connected dominating set problem

The connected dominating set (CDS) problem is a well studied NP-hard problem with many important applications. Dorn et al. (Algorithmica 58:790–810 2010) introduce a branch-decomposition based algorithm design technique for NP-hard problems in planar graphs and give an algorithm (DPBF algorithm) which solves the planar CDS problem in \(O(2^{9.822\sqrt{n}}n+n^3)\) time and \(O(2^{8.11\sqrt{n}}n+n^3)\) time, with a conventional method and fast matrix multiplication in the dynamic programming step of the algorithm, respectively. We show that DPBF algorithm solves the planar CDS problem in \(O(2^{9.8\sqrt{n}}n+n^3)\) time with a conventional method and in \(O(2^{8.08\sqrt{n}}n+n^3)\) time with a fast matrix multiplication. For a graph \(G\), let \({\hbox {bw}}(G)\) be the branchwidth of \(G\) and \(\gamma _c(G)\) be the connected dominating number of \(G\). We prove \({\hbox {bw}}(G)\le 2\sqrt{10\gamma _c(G)}+32\). From this result, the planar CDS problem admits an \(O(2^{23.54\sqrt{\gamma _c(G)}}\gamma _c(G)+n^3)\) time fixed-parameter algorithm. We report computational study results on the practical performance of DPBF algorithm, which show that the size of instances can be solved by the algorithm mainly depends on the branchwidth of the instances, coinciding with the theoretical analysis. For graphs with small or moderate branchwidth, the CDS problem instances with size up to a few thousands edges can be solved in a practical time and memory space.     

Gompertz model with time-dependent covariate in the presence of interval-, right- and left-censored data

In this paper, the Gompertz model is extended to incorporate time-dependent covariates in the presence of interval-, right-, left-censored and uncensored data. Then, its performance at different sample sizes, study periods and attendance probabilities are studied. Following that, the model is compared to a fixed covariate model. Finally, two confidence interval estimation methods, Wald and likelihood ratio (LR), are explored and conclusions are drawn based on the results of the coverage probability study. The results indicate that bias, standard error and root mean square error values of the parameter estimates decrease with the increase in study period, attendance probability and sample size. Also, LR was found to work slightly better than the Wald for parameters of the model.