Nested exposure case-control sampling: a sampling scheme to analyze rare time-dependent exposures

For large cohort studies with rare outcomes, the nested case-control design only requires data collection of small subsets of the individuals at risk. These are typically randomly sampled at the observed event times and a weighted, stratified analysis takes over the role of the full cohort analysis. Motivated by observational studies on the impact of hospital-acquired infection on hospital stay outcome, we are interested in situations, where not necessarily the outcome is rare, but time-dependent exposure such as the occurrence of an adverse event or disease progression is. Using the counting process formulation of general nested case-control designs, we propose three sampling schemes where not all commonly observed outcomes need to be included in the analysis. Rather, inclusion probabilities may be time-dependent and may even depend on the past sampling and exposure history. A bootstrap analysis of a full cohort data set from hospital epidemiology allows us to investigate the practical utility of the proposed sampling schemes in comparison to a full cohort analysis and a too simple application of the nested case-control design, if the outcome is not rare.

     

ETHNIC PLACES, POSTMODERNISM, AND URBAN CHANGE IN HOUSTON

City managers and urban sociologists traditionally viewed ethnic places as undesirable areas of overcrowding and social pathology. Regarded as either transitional districts or obstacles to modernization, ethnic places were subject to slum clearance during the modernist phase of urban development. Ethnic places have acquired a new historical and sentimental salience in the "postmodern" developmental era. Preservationist activists and minority "place entrepreneurs" project ethnic culture and symbolic representations in defensive or proactive ways to stimulate neighborhood revitalization. These trends are evident in the recent urban developmental history of Houston, Texas.     

The Robustness of Trend Stationarity: An Illustration with the Extended Nelson-Plosser Dataset

We re-evaluate Andreu and Spanos's findings in favour of trend stationarity by considering the extended Nelson-Plosser data set. This expanded (to 1988) data set includes a period of rather different behaviour compared with the original Nelson-Plosser data used by Andreou and Spanos. We find that Andreou and Spanos's models (with only minor adjustments) exhibit remarable stability over this extended period, and indicate that their conclusions are more robust than they have shown.     

Estimation of Metabolic Rate Constants in PBPK‐Models from Liver Slice Experiments: What Are the Experimental Needs?

A mechanistic model is presented describing the clearance of a compound in a precision-cut liver slice that is incubated in a culture medium. The problem of estimating metabolic rate constants in PBPK models from liver slice experiments is discussed using identifiability analysis. From the identifiability problem analysis, it appears that in addition to the clearance, the compound's free fraction in the slice and the diffusion rate of the exchange of the compound between culture medium and liver slice should be identified. In addition, knowledge of the culture medium volume, the slice volume, the compound's free fraction, and octanol-water-based partition between medium and slice is presupposed. The formal solution for identification is discussed from the perspective of experimental practice. A formally necessary condition for identification is the sampling of parent compound in liver slice or culture medium. However, due to experimental limitations and errors, sampling the parent compound in the slice together with additional sampling of metabolite pooled from the medium and the slice is required for identification in practice. Moreover, it appears that identification results are unreliable when the value of the intrinsic clearance exceeds the value of the diffusion coefficient, a condition to be verified a posteriori.     

Local spatial data infrastructure, Trujillo-Peru

The development of a Municipal Information System, or currently better known as a local spatial data infrastructure, is considered complex due to the required inter-institutional relationships. In many developing countries Geographical Information Systems (GISs) are introduced but the benefits are modest as no changes take place in technical and organisational structure of organisations. Digital databases and computer-aided design (CAD) maps are mushrooming in great variety within different private and public institutions, municipal organisations and even within single departments and with structures similar to the paper period and thus operating on a stand-alone basis.Many national mapping agencies are not able to provide large-scale digital urban base maps, while the absence or low quality of cadastres makes those basic core data sets unavailable or inaccessible. The result is that duplication and incompatible data are frequently observed and also donor-driven stand-alone projects have a limited impact through the lack of institutional embedding and are not able to mature from the project to the institutional level. However, a positive sign is that there is an increasing awareness among data producers and consumers that investments in the development of digital data sets should be combined to reduce costs and increase benefits from especially GIS, and information and communication technology (ICT) in general.Within Trujillo a long-term vision was developed to make full use of ICT and GIS to modernise all operations of the Municipality to increase the efficiency and effectiveness of its tasks. However, large investments are not feasible due to the very limited municipal budgets. To guarantee the support of the municipal council, short-term results are required. This paper describes three ‘products’ as part of the vision to develop through a step-by-step approach a local spatial data infrastructure for Trujillo.The three, rather different, products are:

1. fiscal cadastre, to increase municipal revenues through property taxation;

2. an ‘environmental atlas’ based on a compatible spatial and attribute data sets from a variety of organisations; and

3. a municipal website with interactive GIS and meta data information.

The paper concludes with some comments on the institutional and political requirements for the successful development of a local spatial data infrastructure and stresses that such an infrastructure is not a product but an incremental development process that will progress only in case such a process is simple, cost effective, user-friendly, and flexible with clear products.Political interest and institutional stability and genuine interest for inter-institutional cooperation will add much to the success of this process.     

On the asymptotic normality of the L1- and L2-errors in histogram density estimation

The L₁ and L₂-errors of the histogram estimate of a density f from a sample X₁,X₂,…,Xn using a cubic partition are shown to be asymptotically normal without any unnecessary conditions imposed on the density f. The asymptotic variances are shown to depend on f only through the corresponding norm of f. From this follows the asymptotic null distribution of a goodness-of-fit test based on the total variation distance, introduced by Györfi and van der Meulen (1991). This note uses the idea of partial inversion for obtaining characteristic functions of conditional distributions, which goes back at least to Bartlett (1938).     

Causal Reasoning from Longitudinal Data*

Abstract. This paper reviews some of the key statistical ideas that are encountered when trying to find empirical support to causal interpretations and conclusions, by applying statistical methods on experimental or observational longitudinal data. In such data, typically a collection of individuals are followed over time, then each one has registered a sequence of covariate measurements along with values of control variables that in the analysis are to be interpreted as causes, and finally the individual outcomes or responses are reported. Particular attention is given to the potentially important problem of confounding. We provide conditions under which, at least in principle, unconfounded estimation of the causal effects can be accomplished. Our approach for dealing with causal problems is entirely probabilistic, and we apply Bayesian ideas and techniques to deal with the corresponding statistical inference. In particular, we use the general framework of marked point processes for setting up the probability models, and consider posterior predictive distributions as providing the natural summary measures for assessing the causal effects. We also draw connections to relevant recent work in this area, notably to Judea Pearl's formulations based on graphical models and his calculus of so‐called do‐probabilities. Two examples illustrating different aspects of causal reasoning are discussed in detail.     

Inside two commercially available statistical expert systems

A decade of research into the applications of artificial intelligence in statistics has finally resulted in the appearance of commercially available statistical expert systems. This paper takes a closer look at two of these systems, which are now commercially available on microcomputers, and shows what knowledge they actually contain and how they operate. It is concluded that although the technology and concepts that drive these systems could still benefit from further improvement, the real challenge lies in defining and constructing the statistical knowledge and strategy that should be incorporated and in presenting the results to the user's full advantage.