This paper uses a “dates as data” approach to understand how grave good use and cemetery space changed across the early medieval period in England. A series of composite kernel density estimations were created, based on a dataset of nearly 1100 graves with associated radiocarbon dates, from between the fourth and ninth centuries AD. This modeling revealed a previously unrecognized peak in grave furnishing around 600 AD, which coincides with a peak in isolated burials, and a low point for unfurnished graves and for small cemeteries. It argues that this peak is unrecognized as previous models of chronological change have focused only on graves containing chronologically distinctive artifacts and highlights the importance of radiocarbon dating as a way of avoiding this limitation.

This article introduces datplot, an R package designed to prepare chronological data for visualization, focusing on the treatment of objects dated to overlapping periods of time. Datplot is suitable for all disciplines in which scientists long for a synoptic method that enables the visualization of the chronology of a collection of heterogeneously dated objects. It is especially helpful for visualizing trends in object assemblages over long periods of time—for example, the development of pottery styles—and it can also assist in the dating of stratigraphy. As both authors come from the field of classical archaeology, the examples and case study demonstrating the functionality of the package analyze classical materials. In particular, we focus on presenting an assemblage of epigraphic evidence from Bithynia (northwestern Turkey), with a microregional focal point in the territory of Nicaea (modern Iznik). In the article, we present the internal methodology of datplot and the process of preparing a dataset of categorically and numerically dated objects. We demonstrate visualizing the data prepared by datplot using kernel density estimation and compare the outcome with more established methods such as histograms and line graphs.

The spatio-temporal dynamics of an outbreak provide important insights to help direct public health resources intended to control transmission. They also provide a focus for detailed epidemiological studies and allow the timing and impact of interventions to be assessed.

A common approach is to aggregate case data to administrative regions. Whilst providing a good visual impression of change over space, this method masks spatial variation and assumes that disease risk is constant across space. Risk factors for COVID-19 (e.g. population density, deprivation and ethnicity) vary from place to place across England so it follows that risk will also vary spatially. Kernel density estimation compares the spatial distribution of cases relative to the underlying population, unfettered by arbitrary geographical boundaries, to produce a continuous estimate of spatially varying risk.

Using test results from healthcare settings in England (Pillar 1 of the UK Government testing strategy) and freely available methods and software, we estimated the spatial and spatio-temporal risk of COVID-19 infection across England for the first 6 months of 2020. Widespread transmission was underway when partial lockdown measures were introduced on 23 March 2020 and the greatest risk erred towards large urban areas. The rapid growth phase of the outbreak coincided with multiple introductions to England from the European mainland. The spatio-temporal risk was highly labile throughout.

In terms of controlling transmission, the most important practical application of our results is the accurate identification of areas within regions that may require tailored intervention strategies. We recommend that this approach is absorbed into routine surveillance outputs in England. Further risk characterisation using widespread community testing (Pillar 2) data is needed as is the increased use of predictive spatial models at fine spatial scales.

Species with specific habitat requirements often remain within their habitat and are characterized by a very sedentary behaviour. We used radio-tracking during a 3-wk campaign to investigate the home ranges and movements of three East African passerine bird species, all adapted to cloud-forest habitats: montane white-eye Zosterops poliogaster mbuluensis, stripe-cheeked greenbul Andropadus milanjensis and white-starred robin Pogonocichla stellata macarthuri. Individuals were observed in a forest-grassland mosaic on top of Chyulu Hills, south Kenya. Based on 15 individuals with a total of 1041 fixes, we found (1) that all three species show a sedentary behaviour restricted to their home forest patch; (2) least activity during the afternoon; and (3) re-colonization into its home patch after translocation into a nearby forest patch. Our findings underline that forest-specialist bird species of the tropics show high site fidelity despite their potential movement ability.

We used satellite telemetry to identify in-water habitat used by individuals in the smallest North-west Atlantic subpopulation of adult nesting loggerhead turtles Caretta caretta during the breeding season. During 2010, 2011 and 2012 breeding periods, a total of 20 adult females used habitats proximal to nesting beaches with various levels of protection within Dry Tortugas National Park. We then used a rapid, high-resolution, digital imaging system to map habitat adjacent to nesting beaches, revealing the diversity and distribution of available benthic cover. Turtle behaviour showing measurable site-fidelity to these diverse habitats has implications for managing protected areas and human activities within them. Protecting diverse benthic areas adjacent to loggerhead turtle nesting beaches here and elsewhere could provide benefits for overall biodiversity conservation.

Digital maps have a large potential to support safety-related Advanced Driver Assistance Systems (ADAS) by providing detailed road and environment information. However, one critical attribute – road accident hotspot – is not available from existing digital maps, and is also difficult to derive from practical surveying. This paper provides a Geographical Information Systems (GIS)-based approach for the production of digital hotspot maps, based on a historical accident dataset and geospatial methods in a GIS. In this approach, firstly the Kernel Density Estimation (KDE) method was used to identify hotspot distribution; secondly the Percent Volume Contour (PVC) method was coupled with KDE to extract hotspot patterns; and finally the map layers of hotspot patterns were integrated with classical navigation maps. Following a description for geospatial hotspot production, the derivation of hotspot property data is also discussed. In order to prove this approach, a small-area case study was carried out in the City Centre of Nottingham. The presented results demonstrate that this approach is useful and effective for solving the hotspot creation problem for ADAS, but other future works will be required to improve data effectiveness.

Identification of high-risk regions of schistosomiasis is important for rational resource allocation and effective control strategies. We conducted the first study to apply the newly developed method of adaptive kernel density estimation (KDE)-based spatial relative risk function (sRRF) to detect the high-risk regions of schistosomiasis in the Guichi region of China and compared it with the fixed KDE-based sRRF. We found that the adaptive KDE-based sRRF had a better ability to depict the heterogeneity of risk regions, but was more sensitive to altering the user-defined smoothing parameters. Specifically, the impact of bandwidths on the estimated risk value and risk significance (P value) was higher for the adaptive KDE-based sRRF, but lower on the estimated risk variation standard error (s.e.) compared with the fixed KDE-based sRRF. Based on this application the adaptive and fixed KDE-based sRRF have their respective advantages and disadvantages and the joint application of the two approaches can warrant the best possible identification of high-risk subregions of diseases.

In this paper, a very useful lemma (in two versions) is proved: itsimplifies notably the essential step to establish a Lindebergcentral limit theorem for dependent processes. Then, applying thislemma to weakly dependent processes introduced in Doukhan andLouhichi (1999), a new central limit theorem is obtained forsample mean or kernel density estimator. Moreover, by using thesubsampling, extensions under weaker assumptions of these centrallimit theorems are provided. All the usual causal or non causaltime series: Gaussian, associated, linear, ARCH(∞),bilinear, Volterra processes, ..., enter this frame.

In performing Bayesian analysis of insurance losses, one usually chooses a parametric conditional loss distribution for each risk and a parametric prior distribution to describe how the conditional distributions vary across the risks. Young (1997) applies techniques from nonparametric density estimation to estimate the prior and uses the estimated model to calculate the predictive mean of future claims given past claims. A shortcoming of this method is that, in estimating the prior, one assumes the average claim amount equals the conditional claim. In this paper, we consider a class of priors obtained by perturbing the one determined nonparametrically, as in Young (1997). We thereby reflect the uncertainty in the prior that arises from the randomness in the claim data. We, then, calculate intervals for the corresponding predictive means. We illustrate our method with data from Dannenburg et al. (1996) and compare the intervals of the predictive means with nonparametric confidence intervals.

To use Bayesian analysis to model insurance losses, one usually chooses a parametric conditional loss distribution for each risk and a parametric prior distribution to describe how the conditional distributions vary across the risks. A criticism of this method is that the prior distribution can be difficult to choose and the resulting model may not represent the loss data very well. In this paper, we apply techniques from nonparametric density estimation to estimate the prior. We use the estimated model to calculate the predictive mean of future claims given past claims. We illustrate our method with simulated data from a mixture of a lognormal conditional over a lognormal prior and find that the estimated predictive mean is more accurate than the linear Bühlmann credibility estimator, even when we use a conditional that is not lognormal.