Forecasting the European Carbon Market

In an effort to meet its obligations under the Kyoto Protocol, in 2005 the European Union introduced a cap-and-trade scheme where mandated installations are allocated permits to emit CO2. Financial markets have developed that allow companies to trade these carbon permits. For the EU to achieve reductions in CO2 emissions at a minimum cost, it is necessary that companies make appropriate investments and policymakers design optimal policies. In an effort to clarify the workings of the carbon market, several recent papers have attempted to statistically model it. However, the European carbon market (EU ETS) has many institutional features that potentially impact on daily carbon prices (and associated nancial futures). As a consequence, the carbon market has properties that are quite different from conventional financial assets traded in mature markets. In this paper, we use dynamic model averaging (DMA) in order to forecast in this newly-developing market. DMA is a recently-developed statistical method which has three advantages over conventional approaches. First, it allows the coefficients on the predictors in a forecasting model to change over time. Second, it allows for the entire fore- casting model to change over time. Third, it surmounts statistical problems which arise from the large number of potential predictors that can explain carbon prices. Our empirical results indicate that there are both important policy and statistical bene ts with our approach. Statistically, we present strong evidence that there is substantial turbulence and change in the EU ETS market, and that DMA can model these features and forecast accurately compared to conventional approaches. From a policy perspective, we discuss the relative and changing role of different price drivers in the EU ETS. Finally, we document the forecast performance of DMA and discuss how this relates to the efficiency and maturity of this market.

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.

Forecasting Inflation Using Dynamic Model Averaging

We forecast quarterly US inflation based on the generalized Phillips curve using econometric methods which incorporate dynamic model averaging. These methods not only allow for coe¢ cients to change over time, but also allow for the entire forecasting model to change over time. We nd that dynamic model averaging leads to substantial forecasting improvements over simple benchmark regressions and more sophisticated approaches such as those using time varying coe¢ cient models. We also provide evidence on which sets of predictors are relevant for forecasting in each period.

A New Model Of Trend Inflation

This paper introduces a new model of trend (or underlying) inflation. In contrast to many earlier approaches, which allow for trend inflation to evolve according to a random walk, ours is a bounded model which ensures that trend inflation is constrained to lie in an interval. The bounds of this interval can either be fixed or estimated from the data. Our model also allows for a time-varying degree of persistence in the transitory component of inflation. The bounds placed on trend inflation mean that standard econometric methods for estimating linear Gaussian state space models cannot be used and we develop a posterior simulation algorithm for estimating the bounded trend inflation model. In an empirical exercise with CPI inflation we find the model to work well, yielding more sensible measures of trend inflation and forecasting better than popular alternatives such as the unobserved components stochastic volatility model.

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.

Time Horizons And Smoothing In The Bank Of England’s Reaction Function: The Contrast Between The Standard GMM And Ex Ante Forecast Approaches

The monetary policy reaction function of the Bank of England is estimated by the standard GMM approach and the ex-ante forecast method developed by Goodhart (2005), with particular attention to the horizons for inflation and output at which each approach gives the best fit. The horizons for the ex-ante approach are much closer to what is implied by the Bank’s view of the transmission mechanism, while the GMM approach produces an implausibly slow adjustment of the interest rate, and suffers from a weak instruments problem. These findings suggest a strong preference for the ex-ante approach.

Time Horizons And Smoothing In The Bank Of England’s Reaction Function: The Contrast Between The Standard GMM And Ex Ante Forecast Approaches

The monetary policy reaction function of the Bank of England is estimated by the standard GMM approach and the ex-ante forecast method developed by Goodhart (2005), with particular attention to the horizons for inflation and output at which each approach gives the best fit. The horizons for the ex-ante approach are much closer to what is implied by the Bank’s view of the transmission mechanism, while the GMM approach produces an implausibly slow adjustment of the interest rate, and suffers from a weak instruments problem. These findings suggest a strong preference for the ex-ante approach.

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.

Using VARs and TVP-VARs with Many Macroeconomic Variables

This paper discusses the challenges faced by the empirical macroeconomist and methods for surmounting them. These challenges arise due to the fact that macroeconometric models potentially include a large number of variables and allow for time variation in parameters. These considerations lead to models which have a large number of parameters to estimate relative to the number of observations. A wide range of approaches are surveyed which aim to overcome the resulting problems. We stress the related themes of prior shrinkage, model averaging and model selection. Subsequently, we consider a particular modelling approach in detail. This involves the use of dynamic model selection methods with large TVP-VARs. A forecasting exercise involving a large US macroeconomic data set illustrates the practicality and empirical success of our approach.

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.

Profile of a serial killer: cellular and molecular approaches to study individual cytotoxic T-cells following therapeutic vaccination.

T-cell vaccination may prevent or treat cancer and infectious diseases, but further progress is required to increase clinical efficacy. Step-by-step improvements of T-cell vaccination in phase I/II clinical studies combined with very detailed analysis of T-cell responses at the single cell level are the strategy of choice for the identification of the most promising vaccine candidates for testing in subsequent large-scale phase III clinical trials. Major aims are to fully identify the most efficient T-cells in anticancer therapy, to characterize their TCRs, and to pinpoint the mechanisms of T-cell recruitment and function in well-defined clinical situations. Here we discuss novel strategies for the assessment of human T-cell responses, revealing in part unprecedented insight into T-cell biology and novel structural principles that govern TCR-pMHC recognition. Together, the described approaches advance our knowledge of T-cell mediated-protection from human diseases.

Time Varying Dimension Models

Time varying parameter (TVP) models have enjoyed an increasing popularity in empirical macroeconomics. However, TVP models are parameter-rich and risk over-fitting unless the dimension of the model is small. Motivated by this worry, this paper proposes several Time Varying dimension (TVD) models where the dimension of the model can change over time, allowing for the model to automatically choose a more parsimonious TVP representation, or to switch between different parsimonious representations. Our TVD models all fall in the category of dynamic mixture models. We discuss the properties of these models and present methods for Bayesian inference. An application involving US inflation forecasting illustrates and compares the different TVD models. We find our TVD approaches exhibit better forecasting performance than several standard benchmarks and shrink towards parsimonious specifications.

Optimisation of antiretroviral therapy : pharmacokinetic and pharmacogenetic approaches

Les progrès de la thérapie antirétrovirale ont transformé l'infection par le VIH d'une condition inévitablement fatale à une maladie chronique. En dépit de ce succès, l'échec thérapeutique et la toxicité médicamenteuse restent fréquents. Une réponse inadéquate au traitement est clairement multifactorielle et une individualisation de la posologie des médicaments qui se baserait sur les facteurs démographiques et génétiques des patients et sur les taux sanguins totaux, libres et/ou cellulaires des médicaments pourrait améliorer à la fois l'efficacité et la tolérance de la thérapie, cette dernière étant certainement un enjeu majeur pour un traitement qui se prend à vie.L'objectif global de cette thèse était de mieux comprendre les facteurs pharmacocinétiques (PK) et pharmacogénétiques (PG) influençant l'exposition aux médicaments antirétroviraux (ARVs) nous offrant ainsi une base rationnelle pour l'optimisation du traitement antiviral et pour l'ajustement posologique des médicaments chez les patients VIH-positifs. Une thérapie antirétrovirale adaptée au patient est susceptible d'augmenter la probabilité d'efficacité et de tolérance à ce traitement, permettant ainsi une meilleure compliance à long terme, et réduisant le risque d'émergence de résistance et d'échec thérapeutique.A cet effet, des méthodes de quantification des concentrations plasmatiques totales, libres et cellulaires des ARVs ainsi que de certains de leurs métabolites ont été développées et validées en utilisant la chromatographie liquide coupée à la spectrométrie de masse en tandem. Ces méthodes ont été appliquées pour la surveillance des taux d'ARVs dans diverses populations de patients HIV-positifs. Une étude clinique a été initiée dans le cadre de l'étude VIH Suisse de cohorte mère-enfant afin de déterminer si la grossesse influence la cinétique des ARVs. Les concentrations totales et libres du lopînavir, de l'atazanavir et de la névirapine ont été déterminées chez les femmes enceintes suivies pendant leur grossesse, et celles-ci ont été trouvées non influencées de manière cliniquement significative par la grossesse. Un ajustement posologique de ces ARVs n'est donc pas nécessaire chez les femmes enceintes. Lors d'une petite étude chez des patients HIV- positifs expérimentés, la corrélation entre l'exposition cellulaire et plasmatique des nouveaux ARVs, notamment le raltégravir, a été déterminée. Une bonne corrélation a été obtenue entre taux plasmatiques et cellulaires de raltégravir, suggérant que la surveillance des taux totaux est un substitut satisfaisant. Cependant, une importante variabilité inter¬patient a été observée dans les ratios d'accumulation cellulaire du raltégravir, ce qui devrait encourager des investigations supplémentaires chez les patients en échec sous ce traitement. L'efficacité du suivi thérapeutique des médicaments (TDM) pour l'adaptation des taux d'efavirenz chez des patients avec des concentrations au-dessus de la cible thérapeutique recommandée a été évaluée lors d'une étude prospective. L'adaptation des doses d'efavirenz basée sur le TDM s'est montrée efficace et sûre, soutenant l'utilisation du TDM chez les patients avec concentrations hors cible thérapeutique. L'impact des polymorphismes génétiques des cytochromes P450 (CYP) 2B6, 2A6 et 3A4/5 sur la pharmacocinétique de l'efavirenz et de ces métabolites a été étudié : un modèle de PK de population intégrant les covariats génétiques et démographiques a été construit. Les variations génétiques fonctionnelles dans les voies de métabolisation principales (CYP2B6) et accessoires {CYP2A6et 3A4/S) de l'efavirenz ont un impact sur sa disposition, et peuvent mener à des expositions extrêmes au médicament. Un? ajustement des doses guidé par le TDM est donc recommandé chez ces patients, en accord avec les polymorphismes génétiques.Ainsi, nous avons démonté qu'en utilisant une approche globale tenant compte à la fois des facteurs PK et PG influençant l'exposition aux ARVs chez les patients infectés, il est possible, si nécessaire, d'individualiser la thérapie antirétrovirale dans des situations diverses. L'optimisation du traitement antirétroviral contribue vraisemblablement à une meilleure efficacité thérapeutique à iong terme tout en réduisant la survenue d'effets indésirables.Résumé grand publicOptimisation de la thérapie antirétrovirale: approches pharmacocinétiques et pharmacogénétiquesLes progrès effectués dans le traitement de l'infection par le virus de llmmunodéficienoe humaine acquise (VIH) ont permis de transformer une affection mortelle en une maladie chronique traitable avec des médicaments de plus en plus efficaces. Malgré ce succès, un certain nombre de patients ne répondent pas de façon optimale à leur traitement etyou souffrent d'effets indésirables médicamenteux entraînant de fréquentes modifications dans leur thérapie. Il a été possible de mettre en évidence que l'efficacité d'un traitement antirétroviral est dans la plupart des cas corrélée aux concentrations de médicaments mesurées dans le sang des patients. Cependant, le virus se réplique dans la cellule, et seule la fraction des médicaments non liée aux protéines du plasma sanguin peut entrer dans la cellule et exercer l'activité antirétrovirale au niveau cellulaire. Il existe par ailleurs une importante variabilité des concentrations sanguines de médicament chez des patients prenant pourtant la même dose de médicament. Cette variabilité peut être due à des facteurs démographiques et/ou génétiques susceptibles d'influencer la réponse au traitement antirétroviral.Cette thèse a eu pour objectif de mieux comprendre les facteurs pharmacologiques et génétiques influençant l'efficacité et ta toxicité des médicaments antirétroviraux, dans le but d'individualiser la thérapie antivirale et d'améliorer le suivi des patients HIV-positifs.A cet effet, des méthodes de dosage très sensibles ont été développées pour permettre la quantification des médicaments antirétroviraux dans le sang et les cellules. Ces méthodes analytiques ont été appliquées dans le cadre de diverses études cliniques réalisées avec des patients. Une des études cliniques a recherché s'il y avait un impact des changements physiologiques liés à la grossesse sur les concentrations des médicaments antirétroviraux. Nous avons ainsi pu démontrer que la grossesse n'influençait pas de façon cliniquement significative le devenir des médicaments antirétroviraux chez les femmes enceintes HIV- positives. La posologie de médicaments ne devrait donc pas être modifiée dans cette population de patientes. Par ailleurs, d'autres études ont portés sur les variations génétiques des patients influençant l'activité enzymatique des protéines impliquées dans le métabolisme des médicaments antirétroviraux. Nous avons également étudié l'utilité d'une surveillance des concentrations de médicament (suivi thérapeutique) dans le sang des patients pour l'individualisation des traitements antiviraux. Il a été possible de mettre en évidence des relations significatives entre l'exposition aux médicaments antirétroviraux et l'existence chez les patients de certaines variations génétiques. Nos analyses ont également permis d'étudier les relations entre les concentrations dans le sang des patients et les taux mesurés dans les cellules où le virus HIV se réplique. De plus, la mesure des taux sanguins de médicaments antirétroviraux et leur interprétation a permis d'ajuster la posologie de médicaments chez les patients de façon efficace et sûre.Ainsi, la complémentarité des connaissances pharmacologiques, génétiques et virales s'inscrit dans l'optique d'une stratégie globale de prise en charge du patient et vise à l'individualisation de la thérapie antirétrovirale en fonction des caractéristiques propres de chaque individu. Cette approche contribue ainsi à l'optimisation du traitement antirétroviral dans la perspective d'un succès du traitement à long terme tout en réduisant la probabilité des effets indésirables rencontrés. - The improvement in antirétroviral therapy has transformed HIV infection from an inevitably fatal condition to a chronic, manageable disease. However, treatment failure and drug toxicity are frequent. Inadequate response to treatment is clearly multifactorial and, therefore, dosage individualisation based on demographic factors, genetic markers and measurement of total, free and/or cellular drug level may increase both drug efficacy and tolerability. Drug tolerability is certainly a major issue for a treatment that must be taken indefinitely.The global objective of this thesis aimed at increasing our current understanding of pharmacokinetic (PK) and pharmacogenetic (PG) factors influencing the exposition to antirétroviral drugs (ARVs) in HIV-positive patients. In turn, this should provide us with a rational basis for antiviral treatment optimisation and drug dosage adjustment in HIV- positive patients. Patient's tailored antirétroviral regimen is likely to enhance treatment effectiveness and tolerability, enabling a better compliance over time, and hence reducing the probability of emergence of viral resistance and treatment failure.To that endeavour, analytical methods for the measurement of total plasma, free and cellular concentrations of ARVs and some of their metabolites have been developed and validated using liquid chromatography coupled with tandem mass spectrometry. These assays have been applied for the monitoring of ARVs levels in various populations of HIV- positive patients. A clinical study has been initiated within the frame of the Mother and Child Swiss HIV Cohort Study to determine whether pregnancy influences the exposition to ARVs. Free and total plasma concentrations of lopinavir, atazanavir and nevirapine have been determined in pregnant women followed during the course of pregnancy, and were found not influenced to a clinically significant extent by pregnancy. Dosage adjustment for these drugs is therefore not required in pregnant women. In a study in treatment- experienced HIV-positive patients, the correlation between cellular and total plasma exposure to new antirétroviral drugs, notably the HIV integrase inhibitor raltegravir, has been determined. A good correlation was obtained between total and cellular levels of raltegravir, suggesting that monitoring of total levels are a satisfactory. However, significant inter-patient variability was observed in raltegravir cell accumulation which should prompt further investigations in patients failing under an integrase inhibitor-based regimen. The effectiveness of therapeutic drug monitoring (TDM) to guide efavirenz dose reduction in patients having concentrations above the recommended therapeutic range was evaluated in a prospective study. TDM-guided dosage adjustment of efavirenz was found feasible and safe, supporting the use of TDM in patients with efavirenz concentrations above therapeutic target. The impact of genetic polymorphisms of cytochromes P450 (CYP) 2B6, 2A6 and 3A4/5 on the PK of efavirenz and its metabolites was studied: a population PK model was built integrating both genetic and demographic covariates. Functional genetic variations in main (CYP2B6) and accessory (2A6, 3A4/5) metabolic pathways of efavirenz have an impact on efavirenz disposition, and may lead to extreme drug exposures. Dosage adjustment guided by TDM is thus required in those patients, according to the pharmacogenetic polymorphism.Thus, we have demonstrated, using a comprehensive approach taking into account both PK and PG factors influencing ARVs exposure in HIV-infected patients, the feasibility of individualising antirétroviral therapy in various situations. Antiviral treatment optimisation is likely to increase long-term treatment success while reducing the occurrence of adverse drug reactions.

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.