Computing probable maximum loss in catastrophe reinsurance portfolios on multi-core and many-core architectures

In the reinsurance market, the risks natural catastrophes pose to portfolios of properties must be quantified, so that they can be priced, and insurance offered. The analysis of such risks at a portfolio level requires a simulation of up to 800 000 trials with an average of 1000 catastrophic events per trial. This is sufficient to capture risk for a global multi-peril reinsurance portfolio covering a range of perils including earthquake, hurricane, tornado, hail, severe thunderstorm, wind storm, storm surge and riverine flooding, and wildfire. Such simulations are both computation and data intensive, making the application of high-performance computing techniques desirable.

In this paper, we explore the design and implementation of portfolio risk analysis on both multi-core and many-core computing platforms. Given a portfolio of property catastrophe insurance treaties, key risk measures, such as probable maximum loss, are computed by taking both primary and secondary uncertainties into account. Primary uncertainty is associated with whether or not an event occurs in a simulated year, while secondary uncertainty captures the uncertainty in the level of loss due to the use of simplified physical models and limitations in the available data. A combination of fast lookup structures, multi-threading and careful hand tuning of numerical operations is required to achieve good performance. Experimental results are reported for multi-core processors and systems using NVIDIA graphics processing unit and Intel Phi many-core accelerators.

Observational Constraints on the Catastrophic Disruption Rate of Small Main Belt Asteroids

We have calculated 90% confidence limits on the steady-state rate of catastrophic disruptions of main belt asteroids in terms of the absolute magnitude at which one catastrophic disruption occurs per year  as a function of the post-disruption increase in brightness (Δm) and subsequent brightness decay rate (τ  ). The confidence limits were calculated using the brightest unknown main belt asteroid (V=18.5) detected with the Pan-STARRS1 (Pan-STARRS1) telescope. We measured the Pan-STARRS1’s catastrophic disruption detection efficiency over a 453-day interval using the Pan-STARRS moving object processing system (MOPS) and a simple model for the catastrophic disruption event’s photometric behavior in a small aperture centered on the catastrophic disruption event. We then calculated the  contours in the ranges from  and  encompassing measured values from known cratering and disruption events and our model’s predictions. Our simplistic catastrophic disruption model suggests that  and  which would imply that H_0≳28—strongly inconsistent withH_{0,B2005=23.26±0.02} predicted by Bottke et al. (Bottke, W.F., Durda, D.D., Nesvorný, D., Jedicke, R., Morbidelli, A., Vokrouhlický, D., Levison, H.F. [2005]. Icarus, 179, 63–94.) using purely collisional models. However, if we assume that H_0=H0,B2005 our results constrain , inconsistent with our simplistic impact-generated catastrophic disruption model. We postulate that the solution to the discrepancy is that >99% of main belt catastrophic disruptions in the size range to which this study was sensitive (∼100 m) are not impact-generated, but are instead due to fainter rotational breakups, of which the recent discoveries of disrupted asteroids P/2013 P5 and P/2013 R3 are probable examples. We estimate that current and upcoming asteroid surveys may discover up to 10 catastrophic disruptions/year brighter than V=18.5.

Perceptions of stressful life events in Northern Irish school children: a longitudinal study.

Adults’ perceptions of stressful life events have been acknowledged as important moderators of the stress adjustment relationship. Until recently, however, there has been a lack of research on children's perceptions of negative life events. This study assesses children's own perceptions of the stressfulness of negative familial, academic and social events as well as events related to the political conflict in Northern Ireland. Method: Developmental changes in children's perceptions of events are traced over time. One hundred and sixty 8-year-old children completed a self-report measure of the perceived stressfulness of a range of negative life events. The sample was drawn from schools in the Greater Belfast area to include children of both genders, primary religious affiliations in Northern Ireland (i.e., Protestant and Roman Catholic) and of varying socio-economic status. Three years later, 113 of these children, then aged 11, were traced through the school system and completed the same measure. Results: Children's perceptions of stressful events are related to a host of social factors. Girls viewed many negative events as more stressful than their male counterparts. Roman Catholic and Protestant children differed in their perceptions of conflict-related events. Perceptions of various types of negative experiences were differentially related to socio-economic status and age. Conclusion: Personal, social and situational factors differentially determine children's perceptions of negative life experiences.

North European last glacial-interglacial transition (LGIT; 15-9 ka) tephrochronology: extended limits and new events

High-precision correlation of palaeoclimatic and palaeoenvironmental records is crucial for testing hypotheses of synchronous change. Although radiocarbon is the traditional method for dating late Quaternary sedimentary sequences, particularly during the last glacial–interglacial transition (LGIT; 15–9?ka), there are inherent problems with the method, particularly during periods of climate change which are often accompanied by major perturbations in atmospheric radiocarbon content. An alternative method is the use of tephras that act as time-parallel marker horizons. Within Europe, numerous volcanic centres are known to have erupted during the LGIT, providing considerable potential for high-precision correlation independent of past radiocarbon fluctuations. Here we report the first identification of the Vedde Ash and Askja Tephra in Ireland, significantly extending the known provenance of these events. We have also identified two new horizons (the Roddans Port Tephras A and B) and tentatively recognise an additional horizon from Vallensgård Mose (Denmark) that provide crucial additional chronological control for the LGIT. Two phases of the Laacher See Tephra (LST) are reported, the lower Laacher See Tephra (LLST) and probably the C2 phase of the Middle Laacher See Tephra (MLST-C2) indicating a more northeasterly distribution of this fan than reported previously.

Testing the timing of radiocarbon-dated events between proxy archives

For interpreting past changes on a regional or global scale, the timings of proxy-inferred events are usually aligned with data from other locations. However, too often chronological uncertainties are ignored in proxy diagrams and multisite comparisons, making it possible for researchers to fall into the trap of sucking separate events into one illusionary event (or vice versa). Here we largely solve this "suck in and smear syndrome" for radiocarbon (14C) dated sequences. In a Bayesian framework, millions of plausible age-models are constructed to quantify the chronological uncertainties within and between proxy archives. We test the technique on replicated high-resolution 14C-dated peat cores deposited during the "Little Ice Age" (c. AD 1400-1900), a period characterized by abrupt climate changes and severe 14C calibration problems. Owing to internal variability in proxy data and uncertainties in age-models, these (and possibly many more) archives are not consistent in recording decadal climate change. Through explicit statistical tests of palaeoenvironmental hypotheses, we can move forward to systematic interpretations of proxy data. However, chronological uncertainties of non-annually resolved palaeoclimate records are too large for answering decadal timescale questions.

Unlinked vital events in census-based longitudinal studies can bias subsequent analysis

Objective: To examine the potential biases arising from the nonlinkage of census records and vital events in longitudinal studies.
Study Design and Setting: A total of 56,396 deaths of residents of Northern Ireland in the 4 years after the 2001 Census were linked to the 2001 Census records. The characteristics of matched and nonmatched death records were compared using multivariate logistic regression. Subject attributes were as recorded on the death certificate.
Results: In total, 3,392 (6.0%) deaths could not be linked to a census record. Linkage rates were lowest in young adults, males, the unmarried, people living in communal establishments, or living in areas that were more deprived or had recorded low census enumeration. For those aged less than 65 years at census, this linkage would exclude from analysis 20.2% of suicides and 19.7% of deaths by external causes.
Conclusion: The nonlinkage of census and death records is a combination of nonenumeration at census and deficient information about the deceased recorded at the time of death. Unmatched individuals may have been more disadvantaged or socially isolated, and analysis based on the linked data set may therefore show some bias and perhaps understate true social gradients.