Inference for clusters of extreme values

Summary. Inference for clusters of extreme values of a time series typically requires the identification of independent clusters of exceedances over a high threshold. The choice of declustering scheme often has a significant effect on estimates of cluster characteristics. We propose an automatic declustering scheme that is justified by an asymptotic result for the times between threshold exceedances. The scheme relies on the extremal index, which we show may be estimated before declustering, and supports a bootstrap procedure for assessing the variability of estimates.     

Politicisation of Conditional Cash Transfers: The Case of Guatemala

This article complements existing studies on the incumbency effects of conditional cash transfers (CCTs) by analysing the principal mechanisms by which Mi Familia Progresa (MIFAPRO) in Guatemala became increasingly politicised through clientelistic vote buying and threats of programme expulsion in case of non‐electoral support. It finds that MIFAPRO constitutes an emblematic case with mixed effects that are particularly relevant in the diffusion of CCTs to other developing countries.     

Former welfare clients in prison: education and self-reported learning problems

The aim of this study is to investigate educational background, educational wishes, actual participation in prison education, and self-reported learning problems among former child welfare clients in prison. A significant concern is that prisoners' educational level is far lower than for the general population. The same concern has also been expressed regarding former child welfare clients. The question is whether prisoners with a background from the child welfare system are especially vulnerable. The survey reported in this paper addressed the entire Norwegian prison population. We distributed a questionnaire containing questions about whether the prisoner or their family had been in contact with the welfare system, educational issues such as educational background, whether they wanted to participate in prison education and actually participated, and to what extent the participants had self-reported learning problems. We also investigated whether such learning problems affected participation in prison education. The results showed that around one-third of the 2065 respondents were former child welfare clients. The mean age of the respondents in the child welfare group was 31 years and 7.9% were women. The results further revealed that those who had been child welfare clients had a lower educational level and more self-reported learning problems than the general prison population, but most of them still wished to start an education while incarcerated.     

A neural' network applied to tlie calculation of lyapunov exponents 1

Chaotic deterministics systems are characterised by the instability of orbits on an attractor. The largest Lyapunov exponent measures on average the exponential growth rate of small deviations along an orbit and gives as such an indication whether or not the dynamic generating process is unstable. The direct method for calculation of the Lyapunov exponent, based on finite differences as formulated by the so-called Wolf-algorithm,fails on medium sized data sets. Alternatively, one can use a neural network with backpropagation to estimate a data generating function. This so-calletl indirect method enables us to recover the theoretical value of the largest Lyapunov exponent in several examples.     

Variations in the pace of old-age mortality decline in seven European countries, 1950–1999: the role of smoking and other factors earlier in life

We examine the variations in the pace of old-age (80+) mortality decline in seven European countries, from 1950 to 1999. Marked variations were found between countries, periods and sexes. While mortality declines were strong in France and England and Wales, modest or no mortality declines were seen in the 1950s and 1960s in the Nordic countries, and since the 1980s in Denmark, the Netherlands and Norway (men only). For non-smoking-related mortality, a high and consistent pace of mortality decline was observed. Mortality decline stagnated among men born between 1890 and 1899, but persisted among women born between 1847 and 1937. The pace of old-age mortality decline correlated with the pace of mortality decline at ages 60–69 among the same cohorts, but only among men and not for non-smoking-related mortality. Smoking, thus, seems more important than other factors originating earlier in life. Our results furthermore indicate substantial future declines in old-age mortality.     

PREDICTION-FOCUSED MODEL SELECTION FOR AUTOREGRESSIVE MODELS

In order to make predictions of future values of a time series, one needs to specify a forecasting model. A popular choice is an autoregressive time‐series model, for which the order of the model is chosen by an information criterion. We propose an extension of the focused information criterion (FIC) for model‐order selection, with emphasis on a high predictive accuracy (i.e. the mean squared forecast error is low). We obtain theoretical results and illustrate by means of a simulation study and some real data examples that the FIC is a valid alternative to the Akaike information criterion (AIC) and the Bayesian information criterion (BIC) for selection of a prediction model. We also illustrate the possibility of using the FIC for purposes other than forecasting, and explore its use in an extended model.     

Energy and Civilization

This paper sketches, very briefly, the theme of ‘Energy and Civilization’ on a far larger canvass. I shall reconsider the well-known fact that, in the course of time, humans have learned to exploit more and more sources of energy. This has required new technologies, new forms of social organization, and new individual skills. In the latter sense, learning to exploit new sources of energy always was also a process of civilization, involving new forms of personal discipline. Starting from what is called “foraging” (when rudimental tools and weapons provided humans with means for using their own physical strength more effectively in order to obtain food, as almost the only source of energy), following and in some ways widening Norbert Elias perspective, the paper analyses the role of fire domestication, agrarianization and industrialization in humanization and civilization processes.     

The small sample performance of estimators of the standard errors of structural equation models

In this paper, we compare five asymptotically, under a correctly specified likelihood, equivalent estimators of the standard errors for parameters in structural equation models. The estimators are evaluated under different conditions regarding (i) sample size, varying between N=50 and 3200, (ii) distributional assumption of the latent variables and the disturbance terms, namely normal, and heavy tailed (t), and (iii) the complexity of the model. For the assessment of the five estimators we use overall performance, relative bias, MSE and coverage of confidence intervals. The analysis reveals substantial differences in the performance of the five asymptotically equal estimators. Most diversity was found for t distributed, i.e. heavy tailed, data.     

Assessment of problem behaviour by foster parents and their foster children

This study investigates the level of agreement between foster parents and foster children about problem behaviour and how this is associated with the breakdown of a foster care placement. The study took a sample of foster parents and their foster children (n= 60), who were aged 11–17 years, and analysed the severity of problem behaviour displayed by the foster children within 3 months of placement in the foster family and a year and a half later. The level of agreement was established with the intra‐class correlation coefficient, which represents both the differences in scores and the differences in ranking. This coefficient shows that there is a statistically significant difference in the level of agreement on the Internalizing Problems scale of the Child Behaviour Checklist between the placements that broke down and those that continued. The level of agreement is associated with the severity score, which in turn is associated with the outcome of the foster care placement. To avoid placement breakdown, problem behaviour should also be assessed from the perspective of the foster child.