Bayesian Model Averaging in the Instrumental Variable Regression Model

This paper considers the instrumental variable regression model when there is uncertainty about the set of instruments, exogeneity restrictions, the validity of identifying restrictions and the set of exogenous regressors. This uncertainty can result in a huge number of models. To avoid statistical problems associated with standard model selection procedures, we develop a reversible jump Markov chain Monte Carlo algorithm that allows us to do Bayesian model averaging. The algorithm is very exible and can be easily adapted to analyze any of the di¤erent priors that have been proposed in the Bayesian instrumental variables literature. We show how to calculate the probability of any relevant restriction (e.g. the posterior probability that over-identifying restrictions hold) and discuss diagnostic checking using the posterior distribution of discrepancy vectors. We illustrate our methods in a returns-to-schooling application.

Purification and activity standardisation of a (166m)Ho solution.

As part of a project to use the long-lived (T(1/2)=1200a) (166m)Ho as reference source in its reference ionisation chamber, IRA standardised a commercially acquired solution of this nuclide using the 4pibeta-gamma coincidence and 4pigamma (NaI) methods. The (166m)Ho solution supplied by Isotope Product Laboratories was measured to have about 5% Europium impurities (3% (154)Eu, 0.94% (152)Eu and 0.9% (155)Eu). Holmium had therefore to be separated from europium, and this was carried out by means of ion-exchange chromatography. The holmium fractions were collected without europium contamination: 162h long HPGe gamma measurements indicated no europium impurity (detection limits of 0.01% for (152)Eu and (154)Eu, and 0.03% for (155)Eu). The primary measurement of the purified (166m)Ho solution with the 4pi (PC) beta-gamma coincidence technique was carried out at three gamma energy settings: a window around the 184.4keV peak and gamma thresholds at 121.8 and 637.3keV. The results show very good self-consistency, and the activity concentration of the solution was evaluated to be 45.640+/-0.098kBq/g (0.21% with k=1). The activity concentration of this solution was also measured by integral counting with a well-type 5''x5'' NaI(Tl) detector and efficiencies computed by Monte Carlo simulations using the GEANT code. These measurements were mutually consistent, while the resulting weighted average of the 4pi NaI(Tl) method was found to agree within 0.15% with the result of the 4pibeta-gamma coincidence technique. An ampoule of this solution and the measured value of the concentration were submitted to the BIPM as a contribution to the Système International de Référence.

Estimation of dynamic latent variable models using simulated nonparametric moments

Abstract. Given a model that can be simulated, conditional moments at a trial parameter value can be calculated with high accuracy by applying kernel smoothing methods to a long simulation. With such conditional moments in hand, standard method of moments techniques can be used to estimate the parameter. Because conditional moments are calculated using kernel smoothing rather than simple averaging, it is not necessary that the model be simulable subject to the conditioning information that is used to define the moment conditions. For this reason, the proposed estimator is applicable to general dynamic latent variable models. It is shown that as the number of simulations diverges, the estimator is consistent and a higher-order expansion reveals the stochastic difference between the infeasible GMM estimator based on the same moment conditions and the simulated version. In particular, we show how to adjust standard errors to account for the simulations. Monte Carlo results show how the estimator may be applied to a range of dynamic latent variable (DLV) models, and that it performs well in comparison to several other estimators that have been proposed for DLV models.

Quantitative integration of geophysical and hydraulic data : from the local towards the regional scale range

Des progrès significatifs ont été réalisés dans le domaine de l'intégration quantitative des données géophysique et hydrologique l'échelle locale. Cependant, l'extension à de plus grandes échelles des approches correspondantes constitue encore un défi majeur. Il est néanmoins extrêmement important de relever ce défi pour développer des modèles fiables de flux des eaux souterraines et de transport de contaminant. Pour résoudre ce problème, j'ai développé une technique d'intégration des données hydrogéophysiques basée sur une procédure bayésienne de simulation séquentielle en deux étapes. Cette procédure vise des problèmes à plus grande échelle. L'objectif est de simuler la distribution d'un paramètre hydraulique cible à partir, d'une part, de mesures d'un paramètre géophysique pertinent qui couvrent l'espace de manière exhaustive, mais avec une faible résolution (spatiale) et, d'autre part, de mesures locales de très haute résolution des mêmes paramètres géophysique et hydraulique. Pour cela, mon algorithme lie dans un premier temps les données géophysiques de faible et de haute résolution à travers une procédure de réduction déchelle. Les données géophysiques régionales réduites sont ensuite reliées au champ du paramètre hydraulique à haute résolution. J'illustre d'abord l'application de cette nouvelle approche dintégration des données à une base de données synthétiques réaliste. Celle-ci est constituée de mesures de conductivité hydraulique et électrique de haute résolution réalisées dans les mêmes forages ainsi que destimations des conductivités électriques obtenues à partir de mesures de tomographic de résistivité électrique (ERT) sur l'ensemble de l'espace. Ces dernières mesures ont une faible résolution spatiale. La viabilité globale de cette méthode est testée en effectuant les simulations de flux et de transport au travers du modèle original du champ de conductivité hydraulique ainsi que du modèle simulé. Les simulations sont alors comparées. Les résultats obtenus indiquent que la procédure dintégration des données proposée permet d'obtenir des estimations de la conductivité en adéquation avec la structure à grande échelle ainsi que des predictions fiables des caractéristiques de transports sur des distances de moyenne à grande échelle. Les résultats correspondant au scénario de terrain indiquent que l'approche d'intégration des données nouvellement mise au point est capable d'appréhender correctement les hétérogénéitées à petite échelle aussi bien que les tendances à gande échelle du champ hydraulique prévalent. Les résultats montrent également une flexibilté remarquable et une robustesse de cette nouvelle approche dintégration des données. De ce fait, elle est susceptible d'être appliquée à un large éventail de données géophysiques et hydrologiques, à toutes les gammes déchelles. Dans la deuxième partie de ma thèse, j'évalue en détail la viabilité du réechantillonnage geostatique séquentiel comme mécanisme de proposition pour les méthodes Markov Chain Monte Carlo (MCMC) appliquées à des probmes inverses géophysiques et hydrologiques de grande dimension . L'objectif est de permettre une quantification plus précise et plus réaliste des incertitudes associées aux modèles obtenus. En considérant une série dexemples de tomographic radar puits à puits, j'étudie deux classes de stratégies de rééchantillonnage spatial en considérant leur habilité à générer efficacement et précisément des réalisations de la distribution postérieure bayésienne. Les résultats obtenus montrent que, malgré sa popularité, le réechantillonnage séquentiel est plutôt inefficace à générer des échantillons postérieurs indépendants pour des études de cas synthétiques réalistes, notamment pour le cas assez communs et importants où il existe de fortes corrélations spatiales entre le modèle et les paramètres. Pour résoudre ce problème, j'ai développé un nouvelle approche de perturbation basée sur une déformation progressive. Cette approche est flexible en ce qui concerne le nombre de paramètres du modèle et lintensité de la perturbation. Par rapport au rééchantillonage séquentiel, cette nouvelle approche s'avère être très efficace pour diminuer le nombre requis d'itérations pour générer des échantillons indépendants à partir de la distribution postérieure bayésienne. - Significant progress has been made with regard to the quantitative integration of geophysical and hydrological data at the local scale. However, extending corresponding approaches beyond the local scale still represents a major challenge, yet is critically important for the development of reliable groundwater flow and contaminant transport models. To address this issue, I have developed a hydrogeophysical data integration technique based on a two-step Bayesian sequential simulation procedure that is specifically targeted towards larger-scale problems. The objective is to simulate the distribution of a target hydraulic parameter based on spatially exhaustive, but poorly resolved, measurements of a pertinent geophysical parameter and locally highly resolved, but spatially sparse, measurements of the considered geophysical and hydraulic parameters. To this end, my algorithm links the low- and high-resolution geophysical data via a downscaling procedure before relating the downscaled regional-scale geophysical data to the high-resolution hydraulic parameter field. I first illustrate the application of this novel data integration approach to a realistic synthetic database consisting of collocated high-resolution borehole measurements of the hydraulic and electrical conductivities and spatially exhaustive, low-resolution electrical conductivity estimates obtained from electrical resistivity tomography (ERT). The overall viability of this method is tested and verified by performing and comparing flow and transport simulations through the original and simulated hydraulic conductivity fields. The corresponding results indicate that the proposed data integration procedure does indeed allow for obtaining faithful estimates of the larger-scale hydraulic conductivity structure and reliable predictions of the transport characteristics over medium- to regional-scale distances. The approach is then applied to a corresponding field scenario consisting of collocated high- resolution measurements of the electrical conductivity, as measured using a cone penetrometer testing (CPT) system, and the hydraulic conductivity, as estimated from electromagnetic flowmeter and slug test measurements, in combination with spatially exhaustive low-resolution electrical conductivity estimates obtained from surface-based electrical resistivity tomography (ERT). The corresponding results indicate that the newly developed data integration approach is indeed capable of adequately capturing both the small-scale heterogeneity as well as the larger-scale trend of the prevailing hydraulic conductivity field. The results also indicate that this novel data integration approach is remarkably flexible and robust and hence can be expected to be applicable to a wide range of geophysical and hydrological data at all scale ranges. In the second part of my thesis, I evaluate in detail the viability of sequential geostatistical resampling as a proposal mechanism for Markov Chain Monte Carlo (MCMC) methods applied to high-dimensional geophysical and hydrological inverse problems in order to allow for a more accurate and realistic quantification of the uncertainty associated with the thus inferred models. Focusing on a series of pertinent crosshole georadar tomographic examples, I investigated two classes of geostatistical resampling strategies with regard to their ability to efficiently and accurately generate independent realizations from the Bayesian posterior distribution. The corresponding results indicate that, despite its popularity, sequential resampling is rather inefficient at drawing independent posterior samples for realistic synthetic case studies, notably for the practically common and important scenario of pronounced spatial correlation between model parameters. To address this issue, I have developed a new gradual-deformation-based perturbation approach, which is flexible with regard to the number of model parameters as well as the perturbation strength. Compared to sequential resampling, this newly proposed approach was proven to be highly effective in decreasing the number of iterations required for drawing independent samples from the Bayesian posterior distribution.

Determination of DNA persistence length by cryo-electron microscopy. Separation of the static and dynamic contributions to the apparent persistence length of DNA.

Axial deflection of DNA molecules in solution results from thermal motion and intrinsic curvature related to the DNA sequence. In order to measure directly the contribution of thermal motion we constructed intrinsically straight DNA molecules and measured their persistence length by cryo-electron microscopy. The persistence length of such intrinsically straight DNA molecules suspended in thin layers of cryo-vitrified solutions is about 80 nm. In order to test our experimental approach, we measured the apparent persistence length of DNA molecules with natural "random" sequences. The result of about 45 nm is consistent with the generally accepted value of the apparent persistence length of natural DNA sequences. By comparing the apparent persistence length to intrinsically straight DNA with that of natural DNA, it is possible to determine both the dynamic and the static contributions to the apparent persistence length.

Steady-state expression of self-regulated genes.

MOTIVATION: Regulatory gene networks contain generic modules such as feedback loops that are essential for the regulation of many biological functions. The study of the stochastic mechanisms of gene regulation is instrumental for the understanding of how cells maintain their expression at levels commensurate with their biological role, as well as to engineer gene expression switches of appropriate behavior. The lack of precise knowledge on the steady-state distribution of gene expression requires the use of Gillespie algorithms and Monte-Carlo approximations. METHODOLOGY: In this study, we provide new exact formulas and efficient numerical algorithms for computing/modeling the steady-state of a class of self-regulated genes, and we use it to model/compute the stochastic expression of a gene of interest in an engineered network introduced in mammalian cells. The behavior of the genetic network is then analyzed experimentally in living cells. RESULTS: Stochastic models often reveal counter-intuitive experimental behaviors, and we find that this genetic architecture displays a unimodal behavior in mammalian cells, which was unexpected given its known bimodal response in unicellular organisms. We provide a molecular rationale for this behavior, and we implement it in the mathematical picture to explain the experimental results obtained from this network.

Standardisation of 18F by a coincidence method using full solid angle detectors.

A solution of (18)F was standardised with a 4pibeta-4pigamma coincidence counting system in which the beta detector is a one-inch diameter cylindrical UPS89 plastic scintillator, positioned at the bottom of a well-type 5''x5'' NaI(Tl) gamma-ray detector. Almost full detection efficiency-which was varied downwards electronically-was achieved in the beta-channel. Aliquots of this (18)F solution were also measured using 4pigamma NaI(Tl) integral counting and Monte Carlo calculated efficiencies as well as the CIEMAT-NIST method. Secondary measurements of the same solution were also performed with an IG11 ionisation chamber whose equivalent activity is traceable to the Système International de Référence through the contribution IRA-METAS made to it in 2001; IRA's degree of equivalence was found to be close to the key comparison reference value (KCRV). The (18)F activity predicted by this coincidence system agrees closely with the ionisation chamber measurement and is compatible within one standard deviation of the other primary measurements. This work demonstrates that our new coincidence system can standardise short-lived radionuclides used in nuclear medicine.

In vivo quantification of neuro-glial metabolism and glial glutamate concentration using 1H-[13C] MRS at 14.1T.

Astrocytes have recently become a major center of interest in neurochemistry with the discoveries on their major role in brain energy metabolism. An interesting way to probe this glial contribution is given by in vivo (13) C NMR spectroscopy coupled with the infusion labeled glial-specific substrate, such as acetate. In this study, we infused alpha-chloralose anesthetized rats with [2-(13) C]acetate and followed the dynamics of the fractional enrichment (FE) in the positions C4 and C3 of glutamate and glutamine with high sensitivity, using (1) H-[(13) C] magnetic resonance spectroscopy (MRS) at 14.1T. Applying a two-compartment mathematical model to the measured time courses yielded a glial tricarboxylic acid (TCA) cycle rate (Vg ) of 0.27 ± 0.02 μmol/g/min and a glutamatergic neurotransmission rate (VNT ) of 0.15 ± 0.01 μmol/g/min. Glial oxidative ATP metabolism thus accounts for 38% of total oxidative metabolism measured by NMR. Pyruvate carboxylase (VPC ) was 0.09 ± 0.01 μmol/g/min, corresponding to 37% of the glial glutamine synthesis rate. The glial and neuronal transmitochondrial fluxes (Vx (g) and Vx (n) ) were of the same order of magnitude as the respective TCA cycle fluxes. In addition, we estimated a glial glutamate pool size of 0.6 ± 0.1 μmol/g. The effect of spectral data quality on the fluxes estimates was analyzed by Monte Carlo simulations. In this (13) C-acetate labeling study, we propose a refined two-compartment analysis of brain energy metabolism based on (13) C turnover curves of acetate, glutamate and glutamine measured with state of the art in vivo dynamic MRS at high magnetic field in rats, enabling a deeper understanding of the specific role of glial cells in brain oxidative metabolism. In addition, the robustness of the metabolic fluxes determination relative to MRS data quality was carefully studied.

Tests for sex-biased dispersal using bi-parentally inherited genetic markers.

Understanding why dispersal is sex-biased in many taxa is still a major concern in evolutionary ecology. Dispersal tends to be male-biased in mammals and female-biased in birds, but counter-examples exist and little is known about sex bias in other taxa. Obtaining accurate measures of dispersal in the field remains a problem. Here we describe and compare several methods for detecting sex-biased dispersal using bi-parentally inherited, codominant genetic markers. If gene flow is restricted among populations, then the genotype of an individual tells something about its origin. Provided that dispersal occurs at the juvenile stage and that sampling is carried out on adults, genotypes sampled from the dispersing sex should on average be less likely (compared to genotypes from the philopatric sex) in the population in which they were sampled. The dispersing sex should be less genetically structured and should present a larger heterozygote deficit. In this study we use computer simulations and a permutation test on four statistics to investigate the conditions under which sex-biased dispersal can be detected. Two tests emerge as fairly powerful. We present results concerning the optimal sampling strategy (varying number of samples, individuals, loci per individual and level of polymorphism) under different amounts of dispersal for each sex. These tests for biases in dispersal are also appropriate for any attribute (e.g. size, colour, status) suspected to influence the probability of dispersal. A windows program carrying out these tests can be freely downloaded from http://www.unil.ch/izea/softwares/fstat.html

A model of cellular dosimetry for macroscopic tumors in radiopharmaceutical therapy.

PURPOSE: In the radiopharmaceutical therapy approach to the fight against cancer, in particular when it comes to translating laboratory results to the clinical setting, modeling has served as an invaluable tool for guidance and for understanding the processes operating at the cellular level and how these relate to macroscopic observables. Tumor control probability (TCP) is the dosimetric end point quantity of choice which relates to experimental and clinical data: it requires knowledge of individual cellular absorbed doses since it depends on the assessment of the treatment's ability to kill each and every cell. Macroscopic tumors, seen in both clinical and experimental studies, contain too many cells to be modeled individually in Monte Carlo simulation; yet, in particular for low ratios of decays to cells, a cell-based model that does not smooth away statistical considerations associated with low activity is a necessity. The authors present here an adaptation of the simple sphere-based model from which cellular level dosimetry for macroscopic tumors and their end point quantities, such as TCP, may be extrapolated more reliably. METHODS: Ten homogenous spheres representing tumors of different sizes were constructed in GEANT4. The radionuclide 131I was randomly allowed to decay for each model size and for seven different ratios of number of decays to number of cells, N(r): 1000, 500, 200, 100, 50, 20, and 10 decays per cell. The deposited energy was collected in radial bins and divided by the bin mass to obtain the average bin absorbed dose. To simulate a cellular model, the number of cells present in each bin was calculated and an absorbed dose attributed to each cell equal to the bin average absorbed dose with a randomly determined adjustment based on a Gaussian probability distribution with a width equal to the statistical uncertainty consistent with the ratio of decays to cells, i.e., equal to Nr-1/2. From dose volume histograms the surviving fraction of cells, equivalent uniform dose (EUD), and TCP for the different scenarios were calculated. Comparably sized spherical models containing individual spherical cells (15 microm diameter) in hexagonal lattices were constructed, and Monte Carlo simulations were executed for all the same previous scenarios. The dosimetric quantities were calculated and compared to the adjusted simple sphere model results. The model was then applied to the Bortezomib-induced enzyme-targeted radiotherapy (BETR) strategy of targeting Epstein-Barr virus (EBV)-expressing cancers. RESULTS: The TCP values were comparable to within 2% between the adjusted simple sphere and full cellular models. Additionally, models were generated for a nonuniform distribution of activity, and results were compared between the adjusted spherical and cellular models with similar comparability. The TCP values from the experimental macroscopic tumor results were consistent with the experimental observations for BETR-treated 1 g EBV-expressing lymphoma tumors in mice. CONCLUSIONS: The adjusted spherical model presented here provides more accurate TCP values than simple spheres, on par with full cellular Monte Carlo simulations while maintaining the simplicity of the simple sphere model. This model provides a basis for complementing and understanding laboratory and clinical results pertaining to radiopharmaceutical therapy.

Multiuser detection in a dynamic environment — Part II: Joint user identification and parameter estimation

The problem of jointly estimating the number, the identities, and the data of active users in a time-varying multiuser environment was examined in a companion paper (IEEE Trans. Information Theory, vol. 53, no. 9, September 2007), at whose core was the use of the theory of finite random sets on countable spaces. Here we extend that theory to encompass the more general problem of estimating unknown continuous parameters of the active-user signals. This problem is solved here by applying the theory of random finite sets constructed on hybrid spaces. We doso deriving Bayesian recursions that describe the evolution withtime of a posteriori densities of the unknown parameters and data.Unlike in the above cited paper, wherein one could evaluate theexact multiuser set posterior density, here the continuous-parameter Bayesian recursions do not admit closed-form expressions. To circumvent this difficulty, we develop numerical approximationsfor the receivers that are based on Sequential Monte Carlo (SMC)methods (“particle filtering”). Simulation results, referring to acode-divisin multiple-access (CDMA) system, are presented toillustrate the theory.

Fusion of data sets in multivariate linear regression with errors-in-variables

We consider the application of normal theory methods to the estimation and testing of a general type of multivariate regressionmodels with errors--in--variables, in the case where various data setsare merged into a single analysis and the observable variables deviatepossibly from normality. The various samples to be merged can differ on the set of observable variables available. We show that there is a convenient way to parameterize the model so that, despite the possiblenon--normality of the data, normal--theory methods yield correct inferencesfor the parameters of interest and for the goodness--of--fit test. Thetheory described encompasses both the functional and structural modelcases, and can be implemented using standard software for structuralequations models, such as LISREL, EQS, LISCOMP, among others. An illustration with Monte Carlo data is presented.

Scaled and adjusted restricted tests in multi-sample analysis of moment structures

We extend to score, Wald and difference test statistics the scaled and adjusted corrections to goodness-of-fit test statistics developed in Satorra and Bentler (1988a,b). The theory is framed in the general context of multisample analysis of moment structures, under general conditions on the distribution of observable variables. Computational issues, as well as the relation of the scaled and corrected statistics to the asymptotic robust ones, is discussed. A Monte Carlo study illustrates thecomparative performance in finite samples of corrected score test statistics.

Three-dimensional patient-specific dosimetry in radioimmunotherapy with 90Y-ibritumomab-tiuxetan.

Aim of the present article was to perform three-dimensional (3D) single photon emission tomography-based dosimetry in radioimmunotherapy (RIT) with (90)Y-ibritumomab-tiuxetan. A custom MATLAB-based code was used to elaborate 3D images and to compare average 3D doses to lesions and to organs at risk (OARs) with those obtained with planar (2D) dosimetry. Our 3D dosimetry procedure was validated through preliminary phantom studies using a body phantom consisting of a lung insert and six spheres with various sizes. In phantom study, the accuracy of dose determination of our imaging protocol decreased when the object volume decreased below 5 mL, approximately. The poorest results were obtained for the 2.58 mL and 1.30 mL spheres where the dose error evaluated on corrected images with regard to the theoretical dose value was -12.97% and -18.69%, respectively. Our 3D dosimetry protocol was subsequently applied on four patients before RIT with (90)Y-ibritumomab-tiuxetan for a total of 5 lesions and 4 OARs (2 livers, 2 spleens). In patient study, without the implementation of volume recovery technique, tumor absorbed doses calculated with the voxel-based approach were systematically lower than those calculated with the planar protocol, with average underestimation of -39% (range from -13.1% to -62.7%). After volume recovery, dose differences reduce significantly, with average deviation of -14.2% (range from -38.7.4% to +3.4%, 1 overestimation, 4 underestimations). Organ dosimetry in one case overestimated, in the other underestimated the dose delivered to liver and spleen. However, both for 2D and 3D approach, absorbed doses to organs per unit administered activity are comparable with most recent literature findings.

Measuring the stability of histogram appearance when the anchor position is changed

Although the histogram is the most widely used density estimator, itis well--known that the appearance of a constructed histogram for a given binwidth can change markedly for different choices of anchor position. In thispaper we construct a stability index $G$ that assesses the potential changesin the appearance of histograms for a given data set and bin width as theanchor position changes. If a particular bin width choice leads to an unstableappearance, the arbitrary choice of any one anchor position is dangerous, anda different bin width should be considered. The index is based on the statisticalroughness of the histogram estimate. We show via Monte Carlo simulation thatdensities with more structure are more likely to lead to histograms withunstable appearance. In addition, ignoring the precision to which the datavalues are provided when choosing the bin width leads to instability. We provideseveral real data examples to illustrate the properties of $G$. Applicationsto other binned density estimators are also discussed.