UK Macroeconomic Forecasting with Many Predictors: Which Models Forecast Best and When Do They Do So?

Block factor methods offer an attractive approach to forecasting with many predictors. These extract the information in these predictors into factors reflecting different blocks of variables (e.g. a price block, a housing block, a financial block, etc.). However, a forecasting model which simply includes all blocks as predictors risks being over-parameterized. Thus, it is desirable to use a methodology which allows for different parsimonious forecasting models to hold at different points in time. In this paper, we use dynamic model averaging and dynamic model selection to achieve this goal. These methods automatically alter the weights attached to different forecasting model as evidence comes in about which has forecast well in the recent past. In an empirical study involving forecasting output and inflation using 139 UK monthly time series variables, we find that the set of predictors changes substantially over time. Furthermore, our results show that dynamic model averaging and model selection can greatly improve forecast performance relative to traditional forecasting methods.

Spatial Interactions in Hedonic Pricing Models: The Urban Housing Market of Aveiro, Portugal

Spatial heterogeneity, spatial dependence and spatial scale constitute key features of spatial analysis of housing markets. However, the common practice of modelling spatial dependence as being generated by spatial interactions through a known spatial weights matrix is often not satisfactory. While existing estimators of spatial weights matrices are based on repeat sales or panel data, this paper takes this approach to a cross-section setting. Specifically, based on an a priori definition of housing submarkets and the assumption of a multifactor model, we develop maximum likelihood methodology to estimate hedonic models that facilitate understanding of both spatial heterogeneity and spatial interactions. The methodology, based on statistical orthogonal factor analysis, is applied to the urban housing market of Aveiro, Portugal at two different spatial scales.

Dimensionality of the Rosenberg self-esteem scale and its relationships with the three- and five-factor personality models

The authors investigated the dimensionality of the French version of the Rosenberg Self-Esteem Scale (RSES; Rosenberg, 1965) using confirmatory factor analysis. We tested models of 1 or 2 factors. Results suggest the RSES is a 1-dimensional scale with 3 highly correlated items. Comparison with the Revised NEO-Personality Inventory (NEO-PI-R; Costa, McCrae, & Rolland, 1998) demonstrated that Neuroticism correlated strongly and Extraversion and Conscientiousness moderately with the RSES. Depression accounted for 47% of the variance of the RSES. Other NEO-PI-R facets were also moderately related with self-esteem.

Measuring the Economic Significance of Structural Exchange Rate Models

This paper examines both the in-sample and out-of-sample performance of three monetary fundamental models of exchange rates and compares their out-of-sample performance to that of a simple Random Walk model. Using a data-set consisting of five currencies at monthly frequency over the period January 1980 to December 2009 and a battery of newly developed performance measures, the paper shows that monetary models do better (in-sample and out-of-sample forecasting) than a simple Random Walk model.

Stability of Growth Models with Generalised Lag Structures

This paper considers the lag structures of dynamic models in economics, arguing that the standard approach is too simple to capture the complexity of actual lag structures arising, for example, from production and investment decisions. It is argued that recent (1990s) developments in the the theory of functional differential equations provide a means to analyse models with generalised lag structures. The stability and asymptotic stability of two growth models with generalised lag structures are analysed. The paper concludes with some speculative discussion of time-varying parameters.

Forward Looking and Myopic Regional Computable General Equilibrium Models. How Significant is the Distinction?

We present a stylized intertemporal forward-looking model able that accommodates key regional economic features, an area where the literature is not well developed. The main difference, from the standard applications, is the role of saving and its implication for the balance of payments. Though maintaining dynamic forward-looking behaviour for agents, the rate of private saving is exogenously determined and so no neoclassical financial adjustment is needed. Also, we focus on the similarities and the differences between myopic and forward-looking models, highlighting the divergences among the main adjustment equations and the resulting simulation outcomes.

No Good Deals - No Bad Models

Faced with the problem of pricing complex contingent claims, an investor seeks to make his valuations robust to model uncertainty. We construct a notion of a model- uncertainty-induced utility function and show that model uncertainty increases the investor's eff ective risk aversion. Using the model-uncertainty-induced utility function, we extend the \No Good Deals" methodology of Cochrane and Sa a-Requejo [2000] to compute lower and upper good deal bounds in the presence of model uncertainty. We illustrate the methodology using some numerical examples.

Identification of particular groups of microRNAs that positively or negatively impact on beta cell function in obese models of type 2 diabetes.

AIMS/HYPOTHESIS: MicroRNAs are key regulators of gene expression involved in health and disease. The goal of our study was to investigate the global changes in beta cell microRNA expression occurring in two models of obesity-associated type 2 diabetes and to assess their potential contribution to the development of the disease. METHODS: MicroRNA profiling of pancreatic islets isolated from prediabetic and diabetic db/db mice and from mice fed a high-fat diet was performed by microarray. The functional impact of the changes in microRNA expression was assessed by reproducing them in vitro in primary rat and human beta cells. RESULTS: MicroRNAs differentially expressed in both models of obesity-associated type 2 diabetes fall into two distinct categories. A group including miR-132, miR-184 and miR-338-3p displays expression changes occurring long before the onset of diabetes. Functional studies indicate that these expression changes have positive effects on beta cell activities and mass. In contrast, modifications in the levels of miR-34a, miR-146a, miR-199a-3p, miR-203, miR-210 and miR-383 primarily occur in diabetic mice and result in increased beta cell apoptosis. These results indicate that obesity and insulin resistance trigger adaptations in the levels of particular microRNAs to allow sustained beta cell function, and that additional microRNA deregulation negatively impacting on insulin-secreting cells may cause beta cell demise and diabetes manifestation. CONCLUSIONS/INTERPRETATION: We propose that maintenance of blood glucose homeostasis or progression toward glucose intolerance and type 2 diabetes may be determined by the balance between expression changes of particular microRNAs.

Environmental Kuznets Curves for Carbon Emissions: A Critical Survey

The empirical finding of an inverse U-shaped relationship between per capita income and pollution, the so-called Environmental Kuznets Curve (EKC), suggests that as countries experience economic growth, environmental deterioration decelerates and thus becomes less of an issue. Focusing on the prime example of carbon emissions, the present article provides a critical review of the new econometric techniques that have questioned the baseline polynomial specification in the EKC literature. We discuss issues related to the functional form, heterogeneity, “spurious” regressions and spatial dependence to address whether and to what extent the EKC can be observed. Despite these new approaches, there is still no clear-cut evidence supporting the existence of the EKC for carbon emissions. JEL classifications: C20; Q32; Q50; O13 Keywords: Environmental Kuznets Curve; Carbon emissions; Functional form; Heterogeneity; “Spurious” regressions; Spatial dependence.Residential satisfaction is often used as a barometer to assess the performance of public policy and programmes designed to raise individuals' well-being. However, the fact that responses elicited from residents might be biased by subjective, non-observable factors casts doubt on whether these responses can be taken as trustable indicators of the individuals' housing situation. Emotional factors such as aspirations or expectations might affect individuals' cognitions of their true residential situation. To disentangle this puzzle, we investigated whether identical residential attributes can be perceived differently depending on tenure status. Our results indicate that tenure status is crucial not only in determining the level of housing satisfaction, but also regarding how dwellers perceive their housing characteristics. Keywords: Housing satisfaction, subjective well-being, homeownership. JEL classification: D1, R2.

Model Switching and Model Averaging in Time- Varying Parameter Regression Models

This paper investigates the usefulness of switching Gaussian state space models as a tool for implementing dynamic model selecting (DMS) or averaging (DMA) in time-varying parameter regression models. DMS methods allow for model switching, where a different model can be chosen at each point in time. Thus, they allow for the explanatory variables in the time-varying parameter regression model to change over time. DMA will carry out model averaging in a time-varying manner. We compare our exact approach to DMA/DMS to a popular existing procedure which relies on the use of forgetting factor approximations. In an application, we use DMS to select different predictors in an in ation forecasting application. We also compare different ways of implementing DMA/DMS and investigate whether they lead to similar results.

Model Uncertainty in Panel Vector Autoregressive Models.

We develop methods for Bayesian model averaging (BMA) or selection (BMS) in Panel Vector Autoregressions (PVARs). Our approach allows us to select between or average over all possible combinations of restricted PVARs where the restrictions involve interdependencies between and heterogeneities across cross-sectional units. The resulting BMA framework can find a parsimonious PVAR specification, thus dealing with overparameterization concerns. We use these methods in an application involving the euro area sovereign debt crisis and show that our methods perform better than alternatives. Our findings contradict a simple view of the sovereign debt crisis which divides the euro zone into groups of core and peripheral countries and worries about financial contagion within the latter group.

Model Uncertainty in Panel Vector Autoregressive Models.

We develop methods for Bayesian model averaging (BMA) or selection (BMS) in Panel Vector Autoregressions (PVARs). Our approach allows us to select between or average over all possible combinations of restricted PVARs where the restrictions involve interdependencies between and heterogeneities across cross-sectional units. The resulting BMA framework can find a parsimonious PVAR specification, thus dealing with overparameterization concerns. We use these methods in an application involving the euro area sovereign debt crisis and show that our methods perform better than alternatives. Our findings contradict a simple view of the sovereign debt crisis which divides the euro zone into groups of core and peripheral countries and worries about financial contagion within the latter group.

The convenient calculation of some test statistics in models of discrete choice

The paper considers the use of artificial regression in calculating different types of score test when the log

Identification and characterization of novel molecular mechanisms involved in initiation and progression of myelination

Schwann cells synthesize a large amount of membrane that form a specialized structure called myelin that surrounds axons and facilitate the transmission of electrical signal along neurons in peripheral nervous system (PNS). Previous studies demonstrated that both Schwann cell differentiation and de-differentiation (in the situation of a nerve injury or demyelinating disease) are regulated by cell-intrinsic regulators including several transcription factors. In particular, the de-differentiation of mature Schwann cells is driven by the activation of multiple negative regulators of myelination including Sox2, c-Jun, Notch and Pax3, all usually expressed in immature Schwann cells and suppressed at the onset of myelination. In order to identify new regulators of myelination involved in the development of the PNS, we analyzed the gene-expression profiling data from developing PNS and from three models of demyelinating neuropathies. This analysis led to the identification of Sox4, a member of the Sox family of transcription factors, as a potential candidate. To characterize the molecular function of Sox4 in PNS, we generated two transgenic lines of mice, which overexpress Sox4 specifically in Schwann cells. Detailed analysis of these mice showed that the overexpression of Sox4 in Schwann cells causes a delay in progression of myelination between post-natal day 2 (P2) and P5. Our in vitro analysis suggested that Sox4 cDNA can be overexpressed while the protein translation is tightly regulated. Interestingly, we observed that Sox4 protein is stabilized in nerves of the CMT4C mouse, a model of the human neuropathy. We therefore crossed Sox4 transgenic mice with CMT4C mice and we observed that Sox4 overexpression exacerbated the neuropathy phenotype in these mice. While recognized as being crucial for the normal function of both neurons and myelinating glial cells, the processes that regulate the beginning of myelination and the nature of the neuro-glial cross-talk remains mostly unknown. In order to gain insight into the molecular pathways involved in the interactions between neurons and associated glial cells, we developed a neuron-glia co-culture system based on microfluidic chambers and successfully induced myelination in this system by ascorbic acid. Importantly, we observed that in addition to acting on Schwann cells, ascorbic acid also modulate neuronal/axonal NRG1/ErbB2-B3 signalling. The experimental setting used in our study thus allowed us to discover a novel phenomena of propagation for myelination in vitro. The further characterization of this event brought us to identify other compounds able to induce myelination: ADAMs secretases inhibitor GM6001 and cyclic-AMP. The results generated during my thesis project are therefore not only important for the advancement of our understanding of how the PNS works, but may also potentially help to develop new therapies aiming at improvement of PNS myelination under disease conditions. - Les cellules de Schwann synthétisent une grande quantité de membrane formant une structure spécialisée appelée myéline qui entoure les axones et facilite la transmission du signal électrique le long des neurones du système nerveux périphérique (SNP). Des études antérieures ont démontré que la différenciation et la dédifférenciation des cellules de Schwann (dans la situation d'une lésion nerveuse ou d'une maladie démyélinisante) sont régulées par des régulateurs cellulaires intrinsèques, incluant plusieurs facteurs de transcription. En particulier, la dédifférenciation des cellules de Schwann matures est contrôlée par l'activation de plusieurs régulateurs négatifs de la myélinisation dont Sox2, c-Jun, Notch et Pax3, tous habituellement exprimés dans des cellules de Schwann immatures et supprimés au début de la myélinisation. Afin d'identifier de nouveaux régulateurs de myélinisation impliqués dans le développement du SNP, nous avons analysé le profil d'expression génique durant le développement du SNP ainsi que dans trois modèles de neuropathies démyélinisantes. Cette analyse a mené à l'identification de Sox4, un membre de la famille des facteurs de transcription Sox, comme étant un candidat potentiel. Dans le but de caractériser la fonction moléculaire de Sox4 dans le SNP, nous avons généré deux lignées transgéniques de souris qui surexpriment Sox4 spécifiquement dans les cellules de Schwann. L'analyse détaillée de ces souris a montré que la surexpression de Sox4 dans les cellules de Schwann provoque un retard dans la progression de la myélinisation entre le jour postnatal 2 (P2) et P5. Notre analyse in vitro a suggéré que l'ADNc de Sox4 peut être surexprimé alors que la traduction des protéines est quand à elle étroitement régulée. De façon intéressante, nous avons observé que la protéine Sox4 est stabilisée dans les nerfs des souris CMT4C, un modèle de neuropathie humaine. Nous avons donc croisé les souris transgéniques Sox4 avec des souris CMT4C et avons observé que la surexpression de Sox4 exacerbe le phénotype de neuropathie chez ces souris. Bien que reconnus comme étant cruciaux pour le fonctionnement normal des neurones et des cellules gliales myélinisantes, les processus qui régulent le début de la myélinisation ainsi que la nature des interactions neurone-glie restent largement méconnus. Afin de mieux comprendre les mécanismes moléculaires impliqués dans les interactions entre les neurones et les cellules gliales leur étant associés, nous avons développé un système de co-culture neurone-glie basé sur des chambres microfluidiques et y avons induit avec succès la myélinisation avec de l'acide ascorbique. Étonnamment, nous avons remarqué que, en plus d'agir sur les cellules de Schwann, l'acide ascorbique module également la voie de signalisation neuronale/axonale NRG1/ErbB2-B3. Le protocole expérimental utilisé dans notre étude a ainsi permis de découvrir un nouveau phénomène de propagation de la myélinisation in vitro. La caractérisation plus poussée de ce phénomène nous a menés à identifier d'autres composés capables d'induire la myélinisation: L'inhibiteur de sécrétases ADAMs GM6001 et l'AMP cyclique. Les résultats obtenus au cours de mon projet de thèse ne sont donc pas seulement importants pour l'avancement de notre compréhension sur la façon dont le SNP fonctionne, mais peuvent aussi potentiellement aider à développer de nouvelles thérapies visant à l'amélioration de la myélinisation du SNP dans des conditions pathologiques.

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.