A bayesian spatial individual effects probit model of the 2010 UK general election

The Conservative Party emerged from the 2010 United Kingdom General Election as the largest single party, but their support was not geographically uniform. In this paper, we estimate a hierarchical Bayesian spatial probit model that tests for the presence of regional voting effects. This model allows for the estimation of individual region-specic effects on the probability of Conservative Party success, incorporating information on the spatial relationships between the regions of the mainland United Kingdom. After controlling for a range of important covariates, we find that these spatial relationships are significant and that our individual region-specic effects estimates provide additional evidence of North-South variations in Conservative Party support.

Bayesian forecasting with highly correlated predictors

This paper considers Bayesian variable selection in regressions with a large number of possibly highly correlated macroeconomic predictors. I show that by acknowledging the correlation structure in the predictors can improve forecasts over existing popular Bayesian variable selection algorithms.

What is the Causal Effect of Information and Learning about a Public Good on Willingness to Pay?

In this study we elicit agents’ prior information set regarding a public good, exogenously give information treatments to survey respondents and subsequently elicit willingness to pay for the good and posterior information sets. The design of this field experiment allows us to perform theoretically motivated hypothesis testing between different updating rules: non-informative updating, Bayesian updating, and incomplete updating. We find causal evidence that agents imperfectly update their information sets. We also field causal evidence that the amount of additional information provided to subjects relative to their pre-existing information levels can affect stated WTP in ways consistent overload from too much learning. This result raises important (though familiar) issues for the use of stated preference methods in policy analysis.

Large Bayesian VARMAs

Vector Autoregressive Moving Average (VARMA) models have many theoretical properties which should make them popular among empirical macroeconomists. However, they are rarely used in practice due to over-parameterization concerns, difficulties in ensuring identification and computational challenges. With the growing interest in multivariate time series models of high dimension, these problems with VARMAs become even more acute, accounting for the dominance of VARs in this field. In this paper, we develop a Bayesian approach for inference in VARMAs which surmounts these problems. It jointly ensures identification and parsimony in the context of an efficient Markov chain Monte Carlo (MCMC) algorithm. We use this approach in a macroeconomic application involving up to twelve dependent variables. We find our algorithm to work successfully and provide insights beyond those provided by VARs.

Limited liability and mechanism design in procurement

In the presence of cost uncertainty, limited liability introduces the possibility of default in procurement with its associated bank-ruptcy costs. When financial soundness is not perfectly observable, we show that incentive compatibility implies that financially less sound contractors are selected with higher probability in any feasible mechanism. Informational rents are associated with unsound financial situations. By selecting the financially weakest contractor, stronger price competition (auctions) may not only increase the probability of default but also expected rents. Thus, weak conditions are suffcient for auctions to be suboptimal. In particular, we show that pooling firms with higher assets may reduce the cost of procurement even when default is costless for the sponsor.

Information Design

There are two ways of creating incentives for interacting agents to behave in a desired way. One is by providing appropriate payoff incentives, which is the subject of mechanism design. The other is by choosing the information that agents observe, which we refer to as information design. We consider a model of symmetric information where a designer chooses and announces the information structure about a payoff relevant state. The interacting agents observe the signal realizations and take actions which affect the welfare of both the designer and the agents. We characterize the general finite approach to deriving the optimal information structure for the designer - the one that maximizes the designer's ex ante expected utility subject to agents playing a Bayes Nash equilibrium. We then apply the general approach to a symmetric two state, two agent, and two actions environment in a parameterized underlying game and fully characterize the optimal information structure: it is never strictly optimal for the designer to use conditionally independent private signals; the optimal information structure may be a public signal or may consist of correlated private signals. Finally, we examine how changes in the underlying game affect the designer's maximum payoff. This exercise provides a joint mechanism/information design perspective.

Design and evaluation of a solid sampler for the monitoring of airborne 1,6-hexamethylene diisocyanate (HDI) and its prepolymers in 2-component spray painting

An active, solvent-free solid sampler was developed for the collection of 1,6-hexamethylene diisocyanate (HDI) aerosol and prepolymers. The sampler was made of a filter impregnated with 1-(2-methoxyphenyl)piperazine contained in a filter holder. Interferences with HDI were observed when a set of cellulose acetate filters and a polystyrene filter holder were used; a glass fiber filter and polypropylene filter cassette gave better results. The applicability of the sampling and analytical procedure was validated with a test chamber, constructed for the dynamic generation of HDI aerosol and prepolymers in commercial two-component spray paints (Desmodur(R) N75) used in car refinishing. The particle size distribution, temporal stability, and spatial uniformity of the simulated aerosol were established in order to test the sample. The monitoring of aerosol concentrations was conducted with the solid sampler paired to the reference impinger technique (impinger flasks contained 10 mL of 0.5 mg/mL 1-(2-methoxyphenyl)piperazine in toluene) under a controlled atmosphere in the test chamber. Analyses of derivatized HDI and prepolymers were carried out by using high-performance liquid chromatography and ultraviolet detection. The correlation between the solvent-free and the impinger techniques appeared fairly good (Y = 0.979X - 0.161; R = 0.978), when the tests were conducted in the range of 0.1 to 10 times the threshold limit value (TLV) for HDI monomer and up to 60-mu-g/m3 (3 U.K. TLVs) for total -N = C = O groups.

Experimental design for acquiring relevant data to address the issue of comparing consecutively manufactured tools and firearms

The comparison of consecutively manufactured tools and firearms has provided much, but not all, of the basis for the profession of firearm and toolmark examination. The authors accept the fundamental soundness of this approach but appeal to the experimental community to close two minor gaps in the experimental procedure. We suggest that "blinding" and attention to appropriateness of other experimental conditions that would consolidate the foundations of our profession. We do not suggest that previous work is unsound.

Bayesian Networks for Probabilistic Inference and Decision Analysis in Forensic Science

Continuing developments in science and technology mean that the amounts of information forensic scientists are able to provide for criminal investigations is ever increasing. The commensurate increase in complexity creates difficulties for scientists and lawyers with regard to evaluation and interpretation, notably with respect to issues of inference and decision. Probability theory, implemented through graphical methods, and specifically Bayesian networks, provides powerful methods to deal with this complexity. Extensions of these methods to elements of decision theory provide further support and assistance to the judicial system. Bayesian Networks for Probabilistic Inference and Decision Analysis in Forensic Science provides a unique and comprehensive introduction to the use of Bayesian decision networks for the evaluation and interpretation of scientific findings in forensic science, and for the support of decision-makers in their scientific and legal tasks. Includes self-contained introductions to probability and decision theory. Develops the characteristics of Bayesian networks, object-oriented Bayesian networks and their extension to decision models. Features implementation of the methodology with reference to commercial and academically available software. Presents standard networks and their extensions that can be easily implemented and that can assist in the reader's own analysis of real cases. Provides a technique for structuring problems and organizing data based on methods and principles of scientific reasoning. Contains a method for the construction of coherent and defensible arguments for the analysis and evaluation of scientific findings and for decisions based on them. Is written in a lucid style, suitable for forensic scientists and lawyers with minimal mathematical background. Includes a foreword by Ian Evett. The clear and accessible style of this second edition makes this book ideal for all forensic scientists, applied statisticians and graduate students wishing to evaluate forensic findings from the perspective of probability and decision analysis. It will also appeal to lawyers and other scientists and professionals interested in the evaluation and interpretation of forensic findings, including decision making based on scientific information.

Parallel spinors on pseudo-riemannian spin(q) manifolds

We study simply-connected irreducible non-locally symmetric pseudo-Riemannian Spin(q) manifolds admitting parallel quaternionic spinors.

New concepts for the design of dermal substitutes

Objectives: Skin can be partially regenerated after full thickness defects by collagen matrices, In this study, we identified the main limitations of induced regeneration aiming to improve the design of dermal matrices. Methods: Single mice received a 1 cm2, full thickness skin wound on the dorsum, which were grafted with collagen-GAG matrices or left ungrafted. The healing modulation induced by the collagen-GAG matrices was compared to spontaneous healing and to custom designed, bioactive, poly-N-Acetyl- Glucosamine (NAG) matrices. Wound staging was based on macroscopic, histological and immunhistochemical analysis on days 3, 7, 10 and 21 post wounding. Results: Cell density was higher in spontaneously granulating wounds compared to grafted wounds. While grafted wounds exhibited increased levels of cell proliferation on days 7 and 10, vascularity was dramatically reduced. NAG scaffolds accelerated both angiogenesis and wound re-epithelialization. Conclusions: Since slow integration and revascularization severely limit the engraftment of clinically used dermal scaffolds, the design of dermal matrices using bioactive materials represent the next step in skin regeneration.

The impact of iron supplementation efficiency in female blood donors with a decreased ferritin level and no anaemia. Rationale and design of a randomised controlled trial: a study protocol.

ABSTRACT: BACKGROUND: There is no recommendation to screen ferritin level in blood donors, even though several studies have noted the high prevalence of iron deficiency after blood donation, particularly among menstruating females. Furthermore, some clinical trials have shown that non-anaemic women with unexplained fatigue may benefit from iron supplementation. Our objective is to determine the clinical effect of iron supplementation on fatigue in female blood donors without anaemia, but with a mean serum ferritin </= 30 ng/ml. METHODS/DESIGN: In a double blind randomised controlled trial, we will measure blood count and ferritin level of women under age 50 yr, who donate blood to the University Hospital of Lausanne Blood Transfusion Department, at the time of the donation and after 1 week. One hundred and forty donors with a ferritin level </= 30 ng/ml and haemoglobin level >/= 120 g/l (non-anaemic) a week after the donation will be included in the study and randomised. A one-month course of oral ferrous sulphate (80 mg/day of elemental iron) will be introduced vs. placebo. Self-reported fatigue will be measured using a visual analogue scale. Secondary outcomes are: score of fatigue (Fatigue Severity Scale), maximal aerobic power (Chester Step Test), quality of life (SF-12), and mood disorders (Prime-MD). Haemoglobin and ferritin concentration will be monitored before and after the intervention. DISCUSSION: Iron deficiency is a potential problem for all blood donors, especially menstruating women. To our knowledge, no other intervention study has yet evaluated the impact of iron supplementation on subjective symptoms after a blood donation. TRIAL REGISTRATION: NCT00689793.

Design and analysis of ChIP-Seq experiments for nuclear factor I DNA-binding proteins

SUMMARY : Eukaryotic DNA interacts with the nuclear proteins using non-covalent ionic interactions. Proteins can recognize specific nucleotide sequences based on the sterical interactions with the DNA and these specific protein-DNA interactions are the basis for many nuclear processes, e.g. gene transcription, chromosomal replication, and recombination. New technology termed ChIP-Seq has been recently developed for the analysis of protein-DNA interactions on a whole genome scale and it is based on immunoprecipitation of chromatin and high-throughput DNA sequencing procedure. ChIP-Seq is a novel technique with a great potential to replace older techniques for mapping of protein-DNA interactions. In this thesis, we bring some new insights into the ChIP-Seq data analysis. First, we point out to some common and so far unknown artifacts of the method. Sequence tag distribution in the genome does not follow uniform distribution and we have found extreme hot-spots of tag accumulation over specific loci in the human and mouse genomes. These artifactual sequence tags accumulations will create false peaks in every ChIP-Seq dataset and we propose different filtering methods to reduce the number of false positives. Next, we propose random sampling as a powerful analytical tool in the ChIP-Seq data analysis that could be used to infer biological knowledge from the massive ChIP-Seq datasets. We created unbiased random sampling algorithm and we used this methodology to reveal some of the important biological properties of Nuclear Factor I DNA binding proteins. Finally, by analyzing the ChIP-Seq data in detail, we revealed that Nuclear Factor I transcription factors mainly act as activators of transcription, and that they are associated with specific chromatin modifications that are markers of open chromatin. We speculate that NFI factors only interact with the DNA wrapped around the nucleosome. We also found multiple loci that indicate possible chromatin barrier activity of NFI proteins, which could suggest the use of NFI binding sequences as chromatin insulators in biotechnology applications. RESUME : L'ADN des eucaryotes interagit avec les protéines nucléaires par des interactions noncovalentes ioniques. Les protéines peuvent reconnaître les séquences nucléotidiques spécifiques basées sur l'interaction stérique avec l'ADN, et des interactions spécifiques contrôlent de nombreux processus nucléaire, p.ex. transcription du gène, la réplication chromosomique, et la recombinaison. Une nouvelle technologie appelée ChIP-Seq a été récemment développée pour l'analyse des interactions protéine-ADN à l'échelle du génome entier et cette approche est basée sur l'immuno-précipitation de la chromatine et sur la procédure de séquençage de l'ADN à haut débit. La nouvelle approche ChIP-Seq a donc un fort potentiel pour remplacer les anciennes techniques de cartographie des interactions protéine-ADN. Dans cette thèse, nous apportons de nouvelles perspectives dans l'analyse des données ChIP-Seq. Tout d'abord, nous avons identifié des artefacts très communs associés à cette méthode qui étaient jusqu'à présent insoupçonnés. La distribution des séquences dans le génome ne suit pas une distribution uniforme et nous avons constaté des positions extrêmes d'accumulation de séquence à des régions spécifiques, des génomes humains et de la souris. Ces accumulations des séquences artéfactuelles créera de faux pics dans toutes les données ChIP-Seq, et nous proposons différentes méthodes de filtrage pour réduire le nombre de faux positifs. Ensuite, nous proposons un nouvel échantillonnage aléatoire comme un outil puissant d'analyse des données ChIP-Seq, ce qui pourraient augmenter l'acquisition de connaissances biologiques à partir des données ChIP-Seq. Nous avons créé un algorithme d'échantillonnage aléatoire et nous avons utilisé cette méthode pour révéler certaines des propriétés biologiques importantes de protéines liant à l'ADN nommés Facteur Nucléaire I (NFI). Enfin, en analysant en détail les données de ChIP-Seq pour la famille de facteurs de transcription nommés Facteur Nucléaire I, nous avons révélé que ces protéines agissent principalement comme des activateurs de transcription, et qu'elles sont associées à des modifications de la chromatine spécifiques qui sont des marqueurs de la chromatine ouverte. Nous pensons que lés facteurs NFI interagir uniquement avec l'ADN enroulé autour du nucléosome. Nous avons également constaté plusieurs régions génomiques qui indiquent une éventuelle activité de barrière chromatinienne des protéines NFI, ce qui pourrait suggérer l'utilisation de séquences de liaison NFI comme séquences isolatrices dans des applications de la biotechnologie.