Windows of Opportunity: Window Subseries Empirical Variance Estimators in International Relations

We show that temporal, spatial, and dyadic dependencies amongobservations complicate the estimation of covariance structuresin panel databases. Ignoring these dependencies results in covarianceestimates that are often too small and inferences that may bemore confident about empirical patterns than is justified bythe data. In this article, we detail the development of a nonparametricapproach, window subseries empirical variance estimators (WSEV),that can more fully capture the impact of these dependencieson the covariance structure. We illustrate this approach ina simulation as well as with a statistical model of internationalconflict similar to many applications in the international relationsliterature.     

Editors' Note

Introduction

Editors' Note     

Compositional variation in bullet lead manufacture

The concentrations of antimony, copper, tin, arsenic, silver, bismuth, and cadmium in lead alloys produced by two smelters and one ammunition manufacturer were determined using inductively coupled plasma-atomic emission spectrometry. These element concentrations were used to measure the variations in composition of lead products that result from various processes involved in the manufacture of lead projectiles. In general, when a pot containing molten lead is used to cast a number of objects, these objects are similar, although not necessarily analytically indistinguishable in their elemental compositions. In each subsequent step in the processing of lead at the smelter and at the ammunition manufacturer, the size of an individual homogeneous melt of lead decreases as more distinct compositions are formed as a result of remelting and mixing of sources, including lead scrap. The ammunition manufacturer in this study produced at least 10 compositionally distinguishable groups of bullet wire in a 19.7-h period. The largest group could potentially be used to produce a maximum of 1.3 million compositionally indistinguishable 40 grain bullets.     

Validation of a 16-locus fluorescent multiplex system

STR multiplexes have been indispensable for the efficient genotyping of forensic samples. The PowerPlex 16 System contains the coreCODIS loci, D3S1358, D5S818, D7S820, D8S1179, D13S317, D16S539, D18S51, D21S11, CSF1PO, FGA, THOI, TPOX, vWA, the sex determinant locus, amelogenin, and two pentanucleotide STR loci, Penta D and Penta E. This multiplex satisfies the locus requirements for most national databases and is the most efficient currently available system due to its single PCR amplification. To provide the groundwork for judicial acceptance, including the publication of primer sequences, and to evaluate laboratory-to-laboratory variation, a developmental validation for casework on this commercially available system was performed in 24 laboratories and produced the following conclusions. Amplification was reliable on a variety of thermal cyclers and product could be analyzed on either an ABI PRISM 310 Genetic Analyzer or an ABI PRISM 377 DNA Sequencer. Genotyping using single source samples was consistent between 0.25 and 2 ng of input DNA template with a few laboratories obtaining complete genotypes at 0.0625 ng. However, heterozygote allele imbalance (<60% peak height balance) caused by stochastic effects was observed at a rate of 13% with 0.125 ng DNA and 22% at 0.0625 ng DNA. Mixture analyses were done using a total of 1 ng of DNA template. Most alleles were detected in mixtures of 4 to 1 and some minor alleles were detected in mixtures of 19 to 1. Optimum amplification cycle number was dependent on the sensitivity of the detection instrument used and could also be adjusted to accommodate larger amounts of DNA on solid supports such as FTA paper. Reaction conditions including volume, annealing temperature, and concentrations of primer, AmpliTaq Gold, and magnesium were shown to be optimal yet robust enough to withstand moderate variations without affecting genotype analysis. Environmental, matrix and standard source analyses revealed an ability to obtain complete genotypes in all sample types except those exposed to 80 degrees C for 12-48 days. Finally, comparison of genotype results from the PowerPlex 16 System with other commercially available systems on non-probative reference and forensic samples showed consistent results.     

Surface sterilization of a maggot using bleach does not interfere with mitochondrial DNA analysis of crop contents

Mitochondrial DNA analysis of a maggot crop can identify the corpse a maggot has been feeding on. Analysis of the crop could be useful in a criminal investigation if maggots are found at a suspected crime scene in the absence of a body, or if there is a question of whether a maggot used in postmortem interval estimations moved onto the corpse from another food source. Such analysis can also resolve a chain of custody dispute if it has been suggested that larval samples have been switched. When recovering DNA from a maggot crop, the analyst must be careful to avoid external contamination. We investigated the effects of three simple wash methods developed to reduce external contamination. Maggots raised on pig liver and intentionally contaminated with cow blood were washed using water, 20% bleach, or a solution containing the enzyme DNase. Only washing the maggots in 20% bleach reduced the amount of vertebrate DNA amplified from the maggots' exteriors. No wash method affected the ability to recover DNA from the maggot crops.     

Analysis of ionic ratios in myocardial tissue and their relation to cardiac damage

The authors evaluated the usefulness of the postmortem biochemical analysis of ionic ratios in different parts of the heart and their relation to cardiac damage caused by chest trauma, as observed by anatomopathologic study. Fifty-nine 59 cases were studied, selected from routine necropsies, and samples were taken from different sites of cardiac tissue. The cause of death was trauma in 40 cases and nontraumatic causes in 19 cases. The object of this study was to analyze the levels of Na+, K+, Ca+2, Mg+2, and Zn+2 in different zones of the heart, and the relationship between intracellular and extracellular ion ratios and the different causes of death and any anatomopathologic alterations observed. The biochemical tests revealed a possible relation between the ionic values and cause of death. Alterations in cell membrane permeability and corresponding modification of the ionic ratios were produced earlier than histologic alterations, which need longer to establish themselves whether or not they follow a traumatic process.     

Diagnostic problems associated with cadaveric trauma from animal activity

Analysis of a series of deaths between 1986 and 2001 resulting from natural disease, accidents, suicides, and homicide, where postmortem animal activity had traumatized bodies, was undertaken at the Forensic Science Center in Adelaide to demonstrate the range of lesions that may occur and problems in interpretation that result. Tissue damage had been caused by a variety of animals, including fly larvae, ants, birds, dogs, rodents, sea lice, and sharks. Postmortem animal activity had disguised injuries, modified wounds, and created the appearances of inflicted injury. Problems with identification occurred after postmortem facial trauma, and loss of organ parenchyma had interfered with, or precluded, the precise determination of the manner of death in some cases. Specific kinds of tissue and organ damage may occur after death, necessitating careful assessment of lesions in a search for characteristic features of animal activity. The pattern of lesions may enable identification of the particular species of animal involved.     

Estimation of body exposure to explosion

An ordnance-disposal expert was killed while disposing of a cache of explosives. The likely position of the body was reconstructed by modeling the explosion as an omnidirectional emission of particles from a model of the explosion site and noting the distribution of particles on a model of a human. The applications and limitations of this method in reconstructing the events and correlation with the injuries noted at autopsy are discussed.