Parsimonious periodic autoregressive models for time series with evolving trend and seasonality

Statistics and Computing - This paper proposes an extension of Periodic AutoRegressive (PAR) modelling for time series with evolving features. The large scale of modern datasets, in fact, implies...     

Una conversación con Michael Billig

Resumen

Este estudio explora el impacto del 11-M en las reacciones emocionales, las formas de afrontamiento y el riesgo percibido en la población española. Han participado 869 personas (el 42,7% cumplimentó el cuestionario antes de los atentados y el 57,3% después). Tras el 11-M aumenta la percepción de riesgo, la intensidad emocional negativa y las conductas de afrontamiento defensivo. Existe una sobreestimación del riesgo personal respecto a la probabilidad real. Los participantes más expuestos a situaciones de riesgo (País Vasco) son los que presentan menor probabilidad de ser víctimas del evento. El miedo induce a adoptar estrategias de tipo defensivo y refuerza la percepción de riesgo mientras que la cólera se asocia a reacciones agresivas. Los resultados confirman que las personas atribuyen respuestas afectivas y formas de afrontamiento más disfuncionales a los otros que a uno mismo. Existe una relativa independencia entre la experiencia directa y la percepción de riesgo. El 11-M no ha repercutido ni en la benevolencia ni en la confianza que se tiene en los otros ni en las creencias de tipo socio-cultural.     

Monitoring Progress in Child Poverty Reduction: Methodological Insights and Illustration to the Case Study of Bangladesh

Important steps have been taken at international summits to set up goals and targets to improve the wellbeing of children worldwide. Now the world also has more and better data to monitor progress. This paper presents a new approach to monitoring progress in child poverty reduction based on the Alkire and Foster adjusted headcount ratio and an array of complementary techniques. A theoretical discussion is accompanied by an assessment of child poverty reduction in Bangladesh based on four rounds of the demographic household survey (1997–2007). Emphasis is given to dimensional monotonicity and decomposability as desirable properties of multidimensional poverty measures. Complementary techniques for analysing changes over time are also illustrated, including the Shapley decomposition of changes in overall poverty, as well as a range of robustness tests and statistical significance tests. The results from Bangladesh illustrate the value added of these new tools and the information they provide for policy. The analysis reveals two paths to multidimensional poverty reduction by either decreasing the incidence of poverty or its intensity, and exposes an uneven distribution of national gains across geographical divisions. The methodology allows an integrated analysis of overall changes yet simultaneously examines progress in each region and in each dimension, retaining the positive features of dashboard approaches. The empirical evidence highlights the need to move beyond the headcount ratio towards new measures of child poverty that reflect the intensity of poverty and multiple deprivations that affect poor children at the same time.     

Deep Learning Models for Health and Safety Risk Prediction in Power Infrastructure Projects

Inappropriate management of health and safety (H&S) risk in power infrastructure projects can result in occupational accidents and equipment damage. Accidents at work have detrimental effects on workers, company, and the general public. Despite the availability of H&S incident data, utilizing them to mitigate accident occurrence effectively is challenging due to inherent limitations of existing data logging methods. In this study, we used a text-mining approach for retrieving meaningful terms from data and develop six deep learning (DL) models for H&S risks management in power infrastructure. The DL models include DNN_classify (risk or no risk), DNN_reg1 (loss time), DNN_reg2 (body injury), DNN_reg3 (plant and fleet), DNN_reg4 (equipment), and DNN_reg5 (environment). An H&S risk database obtained from a leading UK power infrastructure construction company was used in developing the models using the H2O framework of the R language. Performances of DL models were assessed and benchmarked with existing models using test data and appropriate performance metrics. The overall accuracy of the classification model was 0.93. The average R² value for the five regression models was 0.92, with mean absolute error between 0.91 and 0.94. The presented results, in addition to the developed user-interface module, will help practitioners obtain a better understanding of H&S challenges, minimize project costs (such as third-party insurance and equipment repairs), and offer effective strategies to mitigate H&S risk.     

SEPARATING BETWEEN UNOBSERVED CONSUMER TYPES: EVIDENCE FROM AIRLINES

We propose an alternative approach to identify unobserved consumer types and assess whether firms price discriminate. Unlike other screening schemes that rely on quantity discounts or product differentiation, in our finite mixture structure individuals have unit demands and the product is homogeneous. We implement the model using an original U.S. airlines data set. The results support the existence of two demand types. The high‐type “business” traveler is less price sensitive, has a higher valuation, and pays a higher price than the low type “tourist.” The proportion of high types also increases as the departure date nears. (JEL C23, L93, R41)     

Audit committee and internal audit and the quality of earnings: empirical evidence from Spanish companies

Corporate scandals and failures such as Enron, WorldCom, etc. were highly catastrophic and had a terrible effect on stakeholders. The pressure of globalisation and the intense competition have lead to organisations encouraging good corporate governance. Organisations have to promote transparency and accountability of financial information. As Gramling et al. (J Account Lit 23:194–244, 2004) argued there are four important mechanisms of corporate governance include boards of directors, audit committees (among others) and internal and external audit functions. However, this paper will focus on the relationship between two corporate governance mechanisms (audit committee and internal audit function). Thus, our purpose is to analyse the relation between characteristics of the audit committee and internal audit function and earnings management measured as abnormal accruals to test the quality of financial statements. We hypothesise that the association between the effectiveness of the audit committee and the presence of an internal audit function and its relationship with this committee would indicate less opportunity for management to manipulate earnings. Using a sample of 108 non-financial Spanish companies that traded on the Madrid Stock Exchange between 2003 and 2006 (432 observations), we have found that the size and number of meetings of the audit committee had a significant negative association with earnings manipulations. Also, our results suggest that a negative relation between having an internal audit function and earnings management.     

A test for randomness based on a complexity measure

A test for randomness based on a statistic related to the complexity of finite sequences is presented. Simulation of binary sequences under different stochastic models provides estimates of the power of the test. The results show that the test is sensitive to a variety of alternatives to randomness and suggest that the proposed test statistic is a reasonable measure of the stochastic complexity of a finite sequence of discrete random variables.     

A Quantitative Assessment of the Risk for Highly Pathogenic Avian Influenza Introduction into Spain via Legal Trade of Live Poultry

Highly pathogenic avian influenza (HPAI) is considered one of the most important diseases of poultry. During the last 9 years, HPAI epidemics have been reported in Asia, the Americas, Africa, and in 18 countries of the European Union (EU). For that reason, it is possible that the risk for HPAI virus (HPAIV) introduction into Spain may have recently increased. Because of the EU free‐trade policy and because legal trade of live poultry was considered an important route for HPAI spread in certain regions of the world, there are fears that Spain may become HPAIV‐infected as a consequence of the legal introduction of live poultry. However, no quantitative assessment of the risk for HPAIV introduction into Spain or into any other EU member state via the trade of poultry has been published in the peer‐reviewed literature. This article presents the results of the first quantitative assessment of the risk for HPAIV introduction into a free country via legal trade of live poultry, along with estimates of the geographical variation of the risk and of the relative contribution of exporting countries and susceptible poultry species to the risk. The annual mean risk for HPAI introduction into Spain was estimated to be as low as 1.36 × 10⁻³, suggesting that under prevailing conditions, introduction of HPAIV into Spain through the trade of live poultry is unlikely to occur. Moreover, these results support the hypothesis that legal trade of live poultry does not impose a significant risk for the spread of HPAI into EU member states.     

Ein Optimierungsmodell für die wöchentliche Tourenplanung unter Einbeziehung der EU-Sozialvorschriften

The Regulation (EC) No 561/2006 and the Directive 2002/15/EC concern the driving and working hours as well as breaks and rest periods of drivers in road transportation. Although the regulations have an enormous effect on vehicle routing and scheduling, only parts of them have been integrated in few solution approaches and some vehicle routing models so far. This paper starts with the presentation of the restrictions of the relevant European Community regulations. Then, a mixed integer linear programming model for the vehicle routing problem with time windows including all rules of the regulations for a planning horizon of an entire week is presented. The model is solved with CPLEX and the impact of the regulations on the resulting vehicle schedules is analyzed by means of computational experiments.     

MOMENT ESTIMATION IN THE CLASS OF BISEXUAL BRANCHING PROCESSES WITH POPULATION–SIZE DEPENDENT MATING

This paper concerns the estimation of the offspring mean vector, the covariance matrix and the growth rate in the class of bisexual branching processes with population‐size dependent mating. For the proposed estimators, some unconditional moments and some conditioned to non‐extinction are determined and asymptotic properties are established. Confidence intervals are obtained and, as illustration, a simulation example is given.