Swain: Stata module to correct the SEM chi-square overidentification test in small sample sizes or complex models. Statistical Software Components S457617, Boston College Department of Economics.

Swain corrects the chi-square overidentification test (i.e., likelihood ratio test of fit) for structural equation models whethr with or without latent variables. The chi-square statistic is asymptotically correct; however, it does not behave as expected in small samples and/or when the model is complex (cf. Herzog, Boomsma, & Reinecke, 2007). Thus, particularly in situations where the ratio of sample size (n) to the number of parameters estimated (p) is relatively small (i.e., the p to n ratio is large), the chi-square test will tend to overreject correctly specified models. To obtain a closer approximation to the distribution of the chi-square statistic, Swain (1975) developed a correction; this scaling factor, which converges to 1 asymptotically, is multiplied with the chi-square statistic. The correction better approximates the chi-square distribution resulting in more appropriate Type 1 reject error rates (see Herzog & Boomsma, 2009; Herzog, et al., 2007).

Genome organization and gene expression shape the transposable element distribution in the Drosophila melanogaster euchromatin.

The distribution of transposable elements (TEs) in a genome reflects a balance between insertion rate and selection against new insertions. Understanding the distribution of TEs therefore provides insights into the forces shaping the organization of genomes. Past research has shown that TEs tend to accumulate in genomic regions with low gene density and low recombination rate. However, little is known about the factors modulating insertion rates across the genome and their evolutionary significance. One candidate factor is gene expression, which has been suggested to increase local insertion rate by rendering DNA more accessible. We test this hypothesis by comparing the TE density around germline- and soma-expressed genes in the euchromatin of Drosophila melanogaster. Because only insertions that occur in the germline are transmitted to the next generation, we predicted a higher density of TEs around germline-expressed genes than soma-expressed genes. We show that the rate of TE insertions is greater near germline- than soma-expressed genes. However, this effect is partly offset by stronger selection for genome compactness (against excess noncoding DNA) on germline-expressed genes. We also demonstrate that the local genome organization in clusters of coexpressed genes plays a fundamental role in the genomic distribution of TEs. Our analysis shows that-in addition to recombination rate-the distribution of TEs is shaped by the interaction of gene expression and genome organization. The important role of selection for compactness sheds a new light on the role of TEs in genome evolution. Instead of making genomes grow passively, TEs are controlled by the forces shaping genome compactness, most likely linked to the efficiency of gene expression or its complexity and possibly their interaction with mechanisms of TE silencing.

Ocular and systemic bio-distribution of rhodamine-conjugated liposomes loaded with VIP injected into the vitreous of Lewis rats.

PURPOSE: Local delivery of therapeutic molecules encapsulated within liposomes is a promising method to treat ocular inflammation. The purpose of the present study was to define the biodistribution of rhodamine-conjugated liposomes loaded with vasoactive intestinal peptide (VIP), an immunosuppressive neuropeptide, following their intravitreal (IVT) injection in normal rats. METHODS: Healthy seven- to eight-week-old Lewis male rats were injected into the vitreous with empty rhodamine-conjugated liposomes (Rh-Lip) or with VIP-loaded Rh-Lip (VIP-Rh-Lip; 50 mM of lipids with an encapsulation efficiency of 3.0+/-0.4 mmol VIP/mol lipids). Twenty-four h after IVT injection, the eyes, the cervical, mesenteric, and inguinal lymph nodes (LN), and spleen were collected. The phenotype and distribution of cells internalizing Rh-Lip and VIP-Rh-Lip were studied. Determination of VIP expression in ocular tissues and lymphoid organs and interactions with T cells in cervical LN was performed on whole mounted tissues and frozen tissue sections by immunofluorescence and confocal microscopy. RESULTS: In the eye, 24 h following IVT injection, fluorescent liposomes (Rh-Lip and VIP-Rh-Lip) were detected mainly in the posterior segment of the eye (vitreous, inner layer of the retina) and to a lesser extent at the level of the iris root and ciliary body. Liposomes were internalized by activated retinal Müller glial cells, ocular tissue resident macrophages, and rare infiltrating activated macrophages. In addition, fluorescent liposomes were found in the episclera and conjunctiva where free VIP expression was also detected. In lymphoid organs, Rh-Lip and VIP-Rh-Lip were distributed almost exclusively in the cervical lymph nodes (LN) with only a few Rh-Lip-positive cells detected in the spleen and mesenteric LN and none in the inguinal LN. In the cervical LN, Rh-Lip were internalized by resident ED3-positive macrophages adjacent to CD4 and CD8-positive T lymphocytes. Some of these T lymphocytes in close contact with macrophages containing VIP-Rh-Lip expressed VIP. CONCLUSIONS: Liposomes are specifically internalized by retinal Müller glial cells and resident macrophages in the eye. A limited passage of fluorescent liposomes from the vitreous to the spleen via the conventional outflow pathway and the venous circulation was detected. The majority of fluorescent liposomes deposited in the conjunctiva following IVT injection reached the subcapsular sinus of the cervical LN via conjuntival lymphatics. In the cervical LN, Rh-Lip were internalized by resident subcapsular sinus macrophages adjacent to T lymphocytes. Detection of VIP in both macrophages and T cells in cervical LN suggests that IVT injection of VIP-Rh-Lip may increase ocular immune privilege by modulating the loco-regional immune environment. In conclusion, our observations suggest that IVT injection of VIP-loaded liposomes is a promising therapeutic strategy to dampen ocular inflammation by modulating macrophage and T cell activation mainly in the loco-regional immune system.

Drug uptake in a rodent sarcoma model after intravenous injection or isolated lung perfusion of free liposomal doxorubicin /

Rapport de synthèseDrug uptake in a rodent sarcoma model after intravenous injection or isolated lungperfusion of free/liposomal doxorubicinIntroductionLa distribution de doxorubicine libre et doxorubicin liposomale pegylée (Liporubicin?) a été comparée après administration intraveineuse ou application via perfusion isolée du poumon (ILP) dans le parenchyme pulmonaire et dans la tumeur des poumons de rongeurs, porteurs d'une tumeur sarcomateuse.Matériel et méthodeUne tumeur sarcomateuse unique a été générée dans le poumon gauche de 36 rongeurs (Fisher rats) suivie, 10 jours plus tard, par application de doxorubicine ou Liporubicin? soit par perfusion isolée du poumon (n = 20) ou administration intraveineuse (n = 12). Deux différentes concentrations ont été utilisées (100 μg et 400 pg) à doses équimolaires pour les deux formulations de doxorubicine. La concentration des agents cytostatiques ont été mesurées dans la tumeur et le parenchyme pulmonaire à l'aide de chromatographic (HPLC).RésultatsLes résultats indiquent que pour doxorubicine libre, le taux de concentration dans la tumeur et le parenchyme pulmonaire est 3 fois (dosage de 100 μ§) et 10 fois (dosage de 400 plus élevé après ILP par rapport à l'administration intraveineuse. En revanche, pour Liporubicin , le taux de concentration est similaire dans la tumeur et le parenchyme pulmonaire entre ILP et administration intraveineuse, pour les deux doses appliquées.ConclusionPour ILP et administration intraveineuse, le ratio entre accumulation de l'agent cytostatique dans la tumeur versus dans le parenchyme pulmonaire a été comparé pour les deux formulations de doxorubicine ainsi que pour les deux dosages. Pour les deux formulations et dosages de doxorubicine, ILP aboutit à un ratio plus élevé par rapport à l'administration intraveineuse. Cependant, pour les deux formulations et dosages de doxorubicine, ILP résulte également en une distribution de l'agent cytostatique plus hétérogène dans le parenchyme pulmonaire comparé à l'administration intraveineuse.En résumé, l'application de doxorubicine par ILP aboutit donc à une accumulation tumorale élevée et à une augmentation du ratio tumeur-parenchyme pulmonaire, mais en même temps également à une distribution plus hétérogène dans le parenchyme pulmonaire par rapport à l'application intraveineuse. Ceci a été observé pour les deux formulations de doxorubicine et pour les deux dosages appliqué.

The quality of life of disease-free Brazilian breast cancer survivors

The objective of this study was to compare the quality of life of women who survived breast cancer and women who have not had the disease, and to identify associated factors. It is cross-sectional study in which the comparison group includes survivors of breast cancer aged between 40 and 69 years. One hundred and fifty-four women were included, 70 with breast cancer and 84 without the disease. The SF-36 was used to evaluate quality of life. The chi-square test and multivariate linear models were used to compare the groups. The estimated mean physical and mental components were significantly better for the group who survived the disease (51.10 and 52.25, respectively) compared to the group without cancer (47.26 and 47.93, respectively). The study indicates that survivors of breast cancer had a better quality of life compared to women without the disease.

Free Flows, Limited Diversification: Explaining the Fall and Rise of Stock Market Correlations, 1890-2001

Using a new dataset on capital account openness, we investigate why equity return correlations changed over the last century. Based on a new, long-run dataset on capital account regulations in a group of 16 countries over the period 1890-2001, we show that correlations increase as financial markets are liberalized. These findings are robust to controlling for both the Forbes-Rigobon bias and global averages in equity return correlations. We test the robustness of our conclusions, and show that greater synchronization of fundamentals is not the main cause of increasing correlations. These results imply that the home bias puzzle may be smaller than traditionally claimed.

Distribution of neuronal and metabolic markers in the human auditory cortex

SUMMARY The human auditory cortex, located on the supratemporal plane of the temporal lobe, is divided in a primary auditory area and several non-primary areas surrounding it. These different areas show anatomical and functional differences. Many studies have focussed on auditory areas in non-human primates, using investigation techniques such as electrophysiological recordings, tracing of neural connections, or immunohistochemical and histochemical staining. Some of these studies have suggested parallel and hierarchical organization of the cortical auditory areas as well as subcortical auditory relays. In humans, only few studies have investigated these regions immunohistochemically, but activation and lesion studies speak in favour of parallel and hierarchical organization, very similar to that of non-human primates. Calcium-binding proteins and metabolic markers were used to investigate possible correlates of hierarchical and parallel organization in man. Calcium-binding proteins, parvalbumin, calretinin and calbindin, modulate the concentration of intracellular free calcium ions and were found in distinct subpopulations of GABAergic neurons in non-human primates species. In our study, their distribution showed several differences between auditory areas: the primary auditory area was darkly stained for both parvalbumin and calbindin, and their expression rapidly decreased while moving away from the primary area. This staining pattern suggests a hierarchical organization of the areas, in which the more darkly stained areas could correspond to an earlier integration level and the areas showing light staining may correspond to higher level integration areas. Parallel organization of primary and non-primary auditory areas was suggested by the complementarity, within a given area, between parvalbumin and calbindin expression across layers. To investigate the possible differences in the energetic metabolism of the cortical auditory areas, several metabolic markers were used: cytochrome oxidase and LDH1 were used as oxidative metabolism markers and LDH5 was used as glycolytic metabolism marker. The results obtained show a difference in the expression of enzymes involved in oxidative metabolism between areas. In the primary auditory area the oxidative metabolism markers were maximally expressed in layer IV. In contrast, higher order areas showed maximal staining in supragranular layers. The expression of LDH5 varied in patches, but did not differ between the different hierarchical auditory areas. The distribution of the two LDH enzymes isoforms also provides information about cellular aspects of metabolic organization, since neurons expressed the LDH1 isoform whereas astrocytes express primarily LDH5, but some astrocytes also contained the LDH1 isoform. This cellular distribution pattern supports the hypothesis of the existence of an astrocyte-neuron lactate shuttle, previously suggested in rodent studies, and in particular of lactate transfer from astrocytes, which produce lactate from the glucose obtained from the circulation, to neurons that use lactate as energy substrate. In conclusion, the hypothesis of parallel and hierarchical organization of the auditory areas can be supported by CaBPs, cytochrome oxidase and LDH1 distribution. Moreover, the two LDHs cellular distribution pattern support the hypothesis of an astrocyte-neuron lactate shuttle in human cortex.

Bioclimatic constraints to Andean cat distribution: a modelling application for rare species

AimTo identify the bioclimatic niche of the endangered Andean cat (Leopardus jacobita), one of the rarest and least known felids in the world, by developing a species distribution model.LocationSouth America, High Andes and Patagonian steppe. Peru, Bolivia, Chile, Argentina.MethodsWe used 108 Andean cat records to build the models, and 27 to test them, applying the Maxent algorithm to sets of uncorrelated bioclimatic variables from global databases, including elevation. We based our biogeographical interpretations on the examination of the predicted geographic range, the modelled response curves and latitudinal variations in climatic variables associated with the locality data.ResultsSimple bioclimatic models for Andean cats were highly predictive with only 3-4 explanatory variables. The climatic niche of the species was defined by extreme diurnal variations in temperature, cold minimum and moderate maximum temperatures, and aridity, characteristic not only of the Andean highlands but also of the Patagonian steppe. Argentina had the highest representation of suitable climates, and Chile the lowest. The most favourable conditions were centrally located and spanned across international boundaries. Discontinuities in suitable climatic conditions coincided with three biogeographical barriers associated with climatic or topographic transitions.Main conclusionsSimple bioclimatic models can produce useful predictions of suitable climatic conditions for rare species, including major biogeographical constraints. In our study case, these constraints are also known to affect the distribution of other Andean species and the genetic structure of Andean cat populations. We recommend surveys of areas with suitable climates and no Andean cat records, including the corridor connecting two core populations. The inclusion of landscape variables at finer scales, crucially the distribution of Andean cat prey, would contribute to refine our predictions for conservation applications.

Distribution of hospital admissions in a middle-income country of the African region

Introduction: Few studies have reported the distribution of all hospital admissions at the entire country level in low and middle-income countries (LMICs). We examined this question in Seychelles, a rapidly developing small island state in the Africa region, in which access to health care is provided free of charge to all inhabitants through a national health system and all hospital admissions are routinely registered. Methods: Based on all admissions to all hospitals in Seychelles in 2005-2008, we calculated the distribution of hospital admissions, age at admission, length of stay and bed occupancy (i.e. cumulated number of patients * number of days spent in all hospitals) according to both hospital departments and broad causes of diseases (using codes of the ICD-10 classification of diseases). Results: Bed occupancy was largest in the surgical wards (36.7% of all days spent in all hospitals), followed by the medical wards (24.3%), gynecology/obstetrics wards (18.4%), pediatric wards (11.2%), and psychiatric wards (7.2%). According to broad causes of diseases/conditions, bed occupancy was highest for obstetrics/gynecology conditions (19.9% of all days spent at hospital), mental diseases (8.6%), cardiovascular diseases (8.1%), upper aerodigestive/pulmonary diseases (8%), infectious/parasitic diseases (8%), gastrointestinal diseases (7.2%), and urogenital diseases (6.7%). Adjusted to 100'000 population, 153 hospital beds are needed every day, including 31 for obstetrics/gynecologic conditions, 13 for mental diseases, 12 for cardiovascular diseases, 12 for upper aerodigestive diseases, 12 for infectious/parasitic diseases, and 11 for gastrointestinal diseases. Conclusion: Our findings give a good indication of the overall distribution of admissions according to both hospital departments and broad causes of diseases in a middle-income country. These findings provide important information for health care planning at the national level

Clients of sex workers in Switzerland: it makes sense to counsel and propose rapid test for HIV on the street, a preliminary report.

BACKGROUND: Clients of street sex workers may be at higher risk for HIV infection than the general population. Furthermore, there is a lack of knowledge regarding HIV testing of clients of sex workers in developed countries. METHOD: This pilot study assessed the feasibility and acceptance of rapid HIV testing by the clients of street-based sex workers in Lausanne, Switzerland. For 5 evenings, clients in cars were stopped by trained field staff for face-to-face interviews focusing on sex-related HIV risk behaviors and HIV testing history. The clients were then offered a free anonymous rapid HIV test in a bus parked nearby. Rapid HIV testing and counselling were performed by experienced nurse practitioners. Clients with reactive tests were offered confirmatory testing, medical evaluation, and care in our HIV clinic. RESULT: We intercepted 144 men, 112 (77.8%) agreed to be interviewed. Among them, 50 (46.6%) had never been tested for HIV. A total of 31 (27.7%) rapid HIV tests were performed, 16 (51.6%) in clients who had not previously been tested. None were reactive. Initially, 19 (16.9%) additional clients agreed to HIV testing but later declined due to the 40-minute queue for testing. CONCLUSION: This pilot study showed that rapid HIV testing in the red light district of Lausanne was feasible, and that the clients of sex workers accepted testing at an unexpectedly high rate. This setting seems particularly appropriate for targeted HIV screening, since more than 40% of the clients had not previously been tested for HIV even though they engaged in sex-related HIV risk behaviour.

The algebraic equality of two asymptotic tests for the hypothesis that a normal distribution has a specified correlation matrix

It is proved the algebraic equality between Jennrich's (1970) asymptotic$X^2$ test for equality of correlation matrices, and a Wald test statisticderived from Neudecker and Wesselman's (1990) expression of theasymptoticvariance matrix of the sample correlation matrix.

Trade liberalization, exports and technology upgrading: Evidence on the impact of MERCOSUR on Argentinean firms

This paper studies the impact of a regional free trade agreement, MERCOSUR, on technologyupgrading by Argentinean firms. To guide empirical work, I introduce technology choice inMelitz s (2003) model of trade with heterogeneous firms. The joint treatment of the technologyadoption and exporting choices shows that the increase in revenues produced by trade integrationcan induce exporters to upgrade technology. An empirical test of the model reveals that firms inindustries facing higher reductions in Brazil s tariffs increase their investment in technologyfaster. The effect of tariffs on entry in the export market and technology adoption is highest inthe upper-middle range of the firm size distribution, as predicted by the model.

A scaled difference chi-square test statistic for moment structure analysis

A family of scaling corrections aimed to improve the chi-square approximation of goodness-of-fit test statistics in small samples, large models, and nonnormal data was proposed in Satorra and Bentler (1994). For structural equations models, Satorra-Bentler's (SB) scaling corrections are available in standard computer software. Often, however, the interest is not on the overall fit of a model, but on a test of the restrictions that a null model say ${\cal M}_0$ implies on a less restricted one ${\cal M}_1$. If $T_0$ and $T_1$ denote the goodness-of-fit test statistics associated to ${\cal M}_0$ and ${\cal M}_1$, respectively, then typically the difference $T_d = T_0 - T_1$ is used as a chi-square test statistic with degrees of freedom equal to the difference on the number of independent parameters estimated under the models ${\cal M}_0$ and ${\cal M}_1$. As in the case of the goodness-of-fit test, it is of interest to scale the statistic $T_d$ in order to improve its chi-square approximation in realistic, i.e., nonasymptotic and nonnormal, applications. In a recent paper, Satorra (1999) shows that the difference between two Satorra-Bentler scaled test statistics for overall model fit does not yield the correct SB scaled difference test statistic. Satorra developed an expression that permits scaling the difference test statistic, but his formula has some practical limitations, since it requires heavy computations that are notavailable in standard computer software. The purpose of the present paper is to provide an easy way to compute the scaled difference chi-square statistic from the scaled goodness-of-fit test statistics of models ${\cal M}_0$ and ${\cal M}_1$. A Monte Carlo study is provided to illustrate the performance of the competing statistics.

Méthodologie intégrée de microgravimétrie en milieu urbain : test d'application au génie civil sur le métro M2 de Lausanne

RESUME Dès le printemps 2004, la construction d'une 2ème ligne de métro est entreprise dans la ville de Lausanne en Suisse. En reliant Ouchy, au bord du lac Léman (alt. 373 m) à Epalinges (alt. 711 m), le nouveau métro "M2" traversera dès 2008 l'agglomération lausannoise du Sud au Nord sur une distance de 6 km. Depuis l'avant-projet, en 1999, une grande quantité de données géologiques a été récolté et de nombreux forages exécutés sur le site. Ceci nous a donné une occasion unique d'entreprendre une étude de microgravimétrique urbaine de détail. Le mode de creusement du tunnel dépend fortement des matériaux à excaver et il est classiquement du domaine du géologue, avec ses connaissances de la géologie régionale et de la stratigraphie des forages, de fournir à l'ingénieur un modèle géologique. Ce modèle indiquera dans ce cas l'épaisseur des terrains meubles qui recouvrent le soubassement rocheux. La représentativité spatiale d'une information très localisée, comme celle d'un forage, est d'autant plus compliquée que le détail recherché est petit. C'est à ce moment là que la prospection géophysique, plus spécialement gravimétrique, peut apporter des informations complémentaires déterminantes pour régionaliser les données ponctuelles des forages. La microgravimétrie en milieu urbain implique de corriger avec soin les perturbations gravifiques sur la mesure de la pesanteur dues aux effets de la topographie, des bâtiments et des caves afin d'isoler l'effet gravifique dû exclusivement à l'épaisseur du remplissage des terrains meubles. Tenant compte de l'intensité des corrections topographiques en milieu urbain, nous avons donné une grande importance aux sous-sols, leurs effets gravifiques pouvant atteindre l'ordre du dixième de mGal. Nous avons donc intégré ces corrections celle de topographie et traité les effets des bâtiments de manière indépendante. Nous avons inclus dans le modèle numérique de terrain (MNT) la chaussée et les sous-sols afin de construire un modèle numérique de terrain urbain. Nous utiliserons un nouvel acronyme « MNTU »pour décrire ce modèle. Nous proposons d'établir des cartes de corrections topographiques préalables, basées sur les données à disposition fournies par le cadastre en faisant des hypothèses sur la profondeur des sous-sols et la hauteur des bâtiments. Les deux zones de test choisies sont caractéristiques des différents types d'urbanisation présente à Lausanne et se révèlent par conséquent très intéressantes pour élaborer une méthodologie globale de la microgravimétrie urbaine. Le but était d'évaluer l'épaisseur du remplissage morainique sur un fond rocheux molassique se situant à une profondeur variable de quelques mètres à une trentaine de mètres et d'en établir une coupe dans l'axe du futur tracé du métro. Les résultats des modélisations se sont révélés très convaincants en détectant des zones qui diffèrent sensiblement du modèle géologique d'avant projet. Nous avons également démontré que l'application de cette méthode géophysique, non destructive, est à même de limiter le nombre de sondages mécaniques lors de l'avant-projet et du projet définitif, ce qui peut limiter à la fois les coûts et le dérangement engendré par ces travaux de surface. L'adaptabilité de la technique gravimétrique permet d'intervenir dans toutes les différentes phases d'un projet de génie civil comme celui de la construction d'un métro en souterrain. KURZFASSUNG Seit dem Frühling 2004 ist in der Stadt Lausanne (Schweiz) die neue U-Bahn "M2" in Konstruktion. Diese soll auf 6 km Länge die Lausanner Agglomeration von Süd nach Nord durchqueren. Die dem Projekt zu Grunde liegende technische Planung sieht vor, daß die Bahnlinie hauptsächlich in der Molasse angesiedelt sein wird. Seit dem Vorentwurf (1999) ist eine große Anzahl geologischer Angaben gesammelt worden. Daraus ergab sich die einmalige Gelegenheit, die Informationen aus den damit verbundenen zahlreichen Bohrungen zu einer detaillierten mikrogravimetrischen Studie der Stadt Lausanne zu erweitern und zu vervollständigen. Das Ziel bestand darin, die Mächtigkeit der die Molasseüberdeckenden Moräneablagerung abzuschätzen, um eine entsprechendes geologisches Profile entlang der künftigen Bahnlinie zu erstellen. Weiterhin sollte gezeigt werden, daß die Anwendung dieser nicht-invasiven geophysikalischen Methode es ermöglicht, die Anzahl der benötigten Bohrungen sowohl in der Pilotphase wie auch im endgültigen Projekt zu reduzieren, was zu wesentlichen finanziellen Einsparungen in der Ausführung des Werkes beitragen würde. Die beiden in dieser Studie bearbeiteten Testzonen befinden sich im Nordteil und im Stadtzentrum von Lausanne und sind durch eine unterschiedliche Urbanisierung charakterisiert. Das anstehende Gestein liegt in verschiedenen Tiefen: von einigen Metern bis zu etwa dreißig Metern. Diese Zonen weisen alle Schwierigkeiten einer urbanen Bebauung mit hoher Verkehrsdichte auf und waren daher massgebend bei der Ausarbeitung einer globalen mikrogravimetrischen Methodologie für die Stadt Lausanne. Die so entwickelte Technik ermöglicht, die störenden Auswirkungen der Topographie, der Gebäude, der Keller und der Öffentlichen Infrastrukturen sorgfältig zu korrigieren, um so die ausschließlich auf die Mächtigkeit des Lockergesteins zurückzuführenden Effekte zu isolieren. In Bezug auf die Intensität der Auswirkungen der topographischen Korrekturen im Stadtgebiet wurde den Untergeschossen eine besonders grosse Bedeutung zugemessen da die entsprechenden Schwerkrafteffekte eine Grösse von rund einem Zehntel mGal erreichen können. Wir schlagen deshalb vor, vorläufige Karten der topographischen Korrekturen zu erstellen. Diese Korrekturen basieren auf den uns vom Katasterplan gelieferten Daten und einigen Hypothesen bezüglich der Tiefe der Untergeschosse und der Höhe der Gebäude. Die Verfügbarkeit einer derartigen Karte vor der eigentlichen gravimetrischen Messkampagne würde uns erlauben, die Position der Meßstationen besser zu wählen. Wir sahen zudem, daß ein entsprechenden a priori Filter benutzt werden kann, wenn die Form und die Intensität der Anomalie offensichtlich dem entsprechenden Gebäude zugeordnet werden können. Diese Strategie muß jedoch mit Vorsicht angewandt werden, denn falls weitere Anomalien dazukommen, können bedeutende Verschiebungen durch Übèrlagerungen der Schwerewirkung verschiedener Strukturen entstehen. Die Ergebnisse der Modellierung haben sich als sehr überzeugend erwiesen, da sie im Voraus unbekannte sensible Zonen korrekt identifiziert haben. Die Anwendbarkeit der in dieser Arbeit entwickelten gravimetrischen Technik ermöglicht es, während allen Phasen eines Grossbauprojekts, wie zum Beispiel bei der Konstruktion einer unterirdischen U-Bahn, einzugreifen. ABSTRACT Since Spring of 2004 a new metro line has been under construction in the city of Lausanne in Switzerland. The new line, the M2, will be 6 km long and will traverse the city from south to north. The civil engineering project determined that the line would be located primarily in the Molasse. Since the preparatory project in 1999, a great quantity of geological data has been collected, and the many drillings made on the site have proved to be a unique opportunity to undertake a study of urban microgravimetry. The goal was to evaluate the thickness of the morainic filling over the molassic bedrock, and to establish a section along the axis of the future line. It then had to be shown that the application of this nondestructive geophysical method could reduce the number of mechanical surveys required both for a preparatory and a definitive project, which would lead to real savings in the realization of a civil engineering project. The two test zones chosen, one in the northern part of the city and one in the city centre, are characterised by various types of urbanisation. Bedrock is at a depth varying from a few metres to about thirty metres. These zones well exemplify the various difficulties encountered in an urban environment and are therefore very interesting for the development of an overall methodology of urban microgravimetry. Microgravimetry in an urban environment requires careful corrections for gravific disturbances due to the effects of topography, buildings, cellars, and the infrastructure of distribution networks, in order to isolate the gravific effect due exclusively to the thickness of loose soil filling. Bearing in mind the intensity of the topographic corrections in an urban environment, we gave particular importance to basements. Their gravific effects can reach the order of one tenth of one meal, and can influence above all the precision of the Bouguer anomaly. We propose to establish preliminary topographic correction charts based on data provided to us by the land register, by making assumptions on the depths of basements and the heights of buildings. Availability of this chart previous to a gravimetry campaign would enable us to choose optimum measuring sites. We have also seen that an a priori filter can be used when the form and the intensity of the anomaly correspond visually to the corresponding building. This strategy must be used with caution because if other anomalies are to be associated, important shifts can be generated by the superposition of the effects of different structures. The results of the model have proved to be very convincing in detecting previously unknown sensitive zones. The adaptability of the gravimetry technique allows for application in all phases of a civil engineering project such as the construction of an underground metro line. RIASSUNTO Dalla primavera 2004 una nuova linea metropolitana é in costruzione nella città di Losanna in Svizzera. La nuova metropolitana "M2" traverserà per la lunghezza di 6 km il centro urbano di Losanna da sud a nord. II progetto d'ingegneria civile prevedeva un tracciato situato essenzialmente nel fondo roccioso arenaceo terziario (molassa). Dalla redazione del progetto preliminare, avvenuta nel 1999, una grande quantità di dati geologici sono stati raccolti e sono stati eseguiti numerosi sondaggi. Questo sì é presentato come un'occasione unica per mettere a punto uno studio microgravimetrico in ambiente urbano con lo scopo di valutare lo spessore dei terreni sciolti di origine glaciale che ricoprono il fondo roccioso di molassa e di mettere in evidenza come l'applicazione di questo metodo geofisico non distruttivo possa limitare il numero di sondaggi meccanici nella fase di progetto preliminare ed esecutivo con conseguente reale risparmio economico nella realizzazione di una tale opera. Le due zone di test sono situate una nella zona nord e la seconda nel centro storico di Losanna e sono caratterizzate da stili architettonici differenti. II fondo roccioso é situato ad una profondità variabile da qualche metro ad una trentina. Queste due zone sembrano ben rappresentare tutte le difficoltà di un ambiente urbano e ben si prestano per elaborare una metodologia globale per la microgravimetria in ambiente urbano. L'applicazione di questa tecnica nell'ambiente suddetto implica la correzione attenta delle perturbazioni sulla misura dell'accelerazione gravitazionale, causate dalla topografia, gli edifici, le cantine e le infrastrutture dei sottoservizi, per ben isolare il segnale esclusivamente causato dallo spessore dei terreni sciolti. Tenuto conto, dell'intensità delle correzioni topografiche, abbiamo dato grande importanza alle cantine, poiché il loro effetto sulle misure può raggiungere il decimo di mGal. Proponiamo quindi di redigere una carta delle correzioni topografiche preliminare all'acquisizione, facendo delle ipotesi sulla profondità delle cantine e sull'altezza degli edifici, sulla base delle planimetrie catastali. L'analisi di questa carta permetterà di scegliere le posizioni più adatte per le stazioni gravimetriche. Abbiamo anche osservato che un filtro a priori, qualora la forma e l'intensità dell'anomalia fosse facilmente riconducibile in maniera visuale ad un edificio, possa essere efficace. Tuttavia questa strategia deve essere utilizzata con precauzione, poiché può introdurre uno scarto, qualora più anomalie, dovute a differenti strutture, si sovrappongano. I risultati delle modellizzazioni si sono rivelati convincenti, evidenziando zone sensibili non conosciute preventivamente. L'adattabilità della tecnica gravimetrica ha mostrato di poter intervenire in differenti fasi di un progetto di ingegneria civile, quale è quella di un'opera in sotterraneo.

Comparing and validating hypothesis test procedures: Graphical and numerical tools

When the behaviour of a specific hypothesis test statistic is studied by aMonte Carlo experiment, the usual way to describe its quality is by givingthe empirical level of the test. As an alternative to this procedure, we usethe empirical distribution of the obtained \emph{p-}values and exploit itsinformation both graphically and numerically.