On the Sources of Uncertainty in Exchange Rate Predictability

We analyse the role of time-variation in coefficients and other sources of uncertainty in exchange rate forecasting regressions. Our techniques incorporate the notion that the relevant set of predictors and their corresponding weights, change over time. We find that predictive models which allow for sudden rather than smooth, changes in coefficients significantly beat the random walk benchmark in out-of-sample forecasting exercise. Using innovative variance decomposition scheme, we identify uncertainty in coefficients' estimation and uncertainty about the precise degree of coefficients' variability, as the main factors hindering models' forecasting performance. The uncertainty regarding the choice of the predictor is small.

Effect of training-sample size and classification difficulty on the accuracy of genomic predictors.

Introduction: As part of the MicroArray Quality Control (MAQC)-II project, this analysis examines how the choice of univariate feature-selection methods and classification algorithms may influence the performance of genomic predictors under varying degrees of prediction difficulty represented by three clinically relevant endpoints. Methods: We used gene-expression data from 230 breast cancers (grouped into training and independent validation sets), and we examined 40 predictors (five univariate feature-selection methods combined with eight different classifiers) for each of the three endpoints. Their classification performance was estimated on the training set by using two different resampling methods and compared with the accuracy observed in the independent validation set. Results: A ranking of the three classification problems was obtained, and the performance of 120 models was estimated and assessed on an independent validation set. The bootstrapping estimates were closer to the validation performance than were the cross-validation estimates. The required sample size for each endpoint was estimated, and both gene-level and pathway-level analyses were performed on the obtained models. Conclusions: We showed that genomic predictor accuracy is determined largely by an interplay between sample size and classification difficulty. Variations on univariate feature-selection methods and choice of classification algorithm have only a modest impact on predictor performance, and several statistically equally good predictors can be developed for any given classification problem.

Comparing population distributions from bin-aggregated sample data: an application to historical height data from France

This paper develops a methodology to estimate the entire population distributions from bin-aggregated sample data. We do this through the estimation of the parameters of mixtures of distributions that allow for maximal parametric flexibility. The statistical approach we develop enables comparisons of the full distributions of height data from potential army conscripts across France's 88 departments for most of the nineteenth century. These comparisons are made by testing for differences-of-means stochastic dominance. Corrections for possible measurement errors are also devised by taking advantage of the richness of the data sets. Our methodology is of interest to researchers working on historical as well as contemporary bin-aggregated or histogram-type data, something that is still widely done since much of the information that is publicly available is in that form, often due to restrictions due to political sensitivity and/or confidentiality concerns.

System GMM estimation with a small sample

Properties of GMM estimators for panel data, which have become very popular in the empirical economic growth literature, are not well known when the number of individuals is small. This paper analyses through Monte Carlo simulations the properties of various GMM and other estimators when the number of individuals is the one typically available in country growth studies. It is found that, provided that some persistency is present in the series, the system GMM estimator has a lower bias and higher efficiency than all the other estimators analysed, including the standard first-differences GMM estimator.

Spectral analysis of random sparse matrices

"Vegeu el resum a l'inici del document del fitxer adjunt."

Propagation connectivity of random hypergraphs

We study the concept of propagation connectivity on random 3-uniform hypergraphs. This concept is inspired by a simple linear time algorithm for solving instances of certain constraint satisfaction problems. We derive upper and lower bounds for the propagation connectivity threshold, and point out some algorithmic implications.

Electrocardiographic alteration among first degree relatives with serologic evidence of Trypanosoma cruzi infection: a sibship study

To analyze whether electrocardiographic alterations (ECGA) in patients with antibodies to Trypanosoma cruzi showed a patttern of familial aggregation, a sample of 379 young adults (166 men and 213 women) distributed in sibships, were assessed for the presence of anti-T.cruzi antibodies, and subjected to a complete clinical examination and a standard resting electrocardiogram (ECG). Positive T. cruzi serology was detected in 165 individuals, 48 of them showing an abnormal ECG (overall prevalence 29 por cento). One hundred and eleven seropositive individuals were distributed in 45 sibships, each of them constituted by more than one seropositive sib, with ECGA being present in 34 out of these patients. Seropositive subjects with ECGA were detected in 27 sibships. Since the index case within each sibship is counted exactly once, affected individuals selected at random as propositi were extracted to calculate the prevalence of ECGA among first degree relatives of probands. Abnormal ECGs were recorded in 7 out of 45 sibs yielding a prevalence that did not differ from estimations registered in the general population or seropositive sibs. Data from the present sample show no familial aggregation for the occurrence of ECGA in patients with T.cruzi infection.

Sequencing and identification of expressed Schistosoma mansoni genes by random selection of cDNA clones from a directional library

We have initiated a gene discovery program in Schistosoma mansoni based on the technique of Expressed Sequence Tags (ESTs), i.e. partial sequences of cDNAs obtained from single passes in automatic DNA sequencers. ESTs can be used to identify genese onf the basis of their homology whith sequences from other species deposited in DNA or protein databases. Trasncripts with sequences without matches in teh databases may represent novel parasite-specific genes. This approach has shown to be very efficient and in less than two years a broad range of novel genes has already been ascertained, more than doubling the number of known S. mansoni genes.

Externalizing disorders and substance use: empirically derived subtypes in a population-based sample of adults.

PURPOSE: Attention-deficit/hyperactivity disorder (ADHD), conduct disorder (CD), and oppositional defiant disorder (ODD) are common externalizing disorders of childhood. The common effects of these disorders on substance abuse need further investigation. The current study investigated the joint clusters of childhood/adolescence ADHD, CD, and ODD, and their influence on substance abuse/dependence in a population-based sample of adults. METHODS: The data were drawn from the PsyCoLaus study (n = 3,720) conducted in Lausanne, Switzerland. The population-based sample included 238 subjects meeting criteria for ADHD/ODD/CD diagnoses before the age of 15. Latent class analyses (LCA) were performed to derive comorbidity subtypes, which were subsequently characterized with respect to psychosocial correlates and substance use. RESULTS: The best fit in LCAs was achieved with three latent classes: an ADHD subtype (35.7 %); an externalizing multimorbid subtype (33.6 %) involving ODD, ADHD, and CD; and a third subtype with CD (30.7 %). The CD subtype showed the highest association with substance use. Apart from this, the externalizing multimorbid subtype was also significantly linked to substance use. The ADHD subtype had only elevated frequencies for alcohol dependence in comparison with subjects that had no history of ADHD, ODD, and CD during childhood or adolescence. Finally, important interactions between subtypes and sex were observed with regard to substance use. CONCLUSIONS: This study provides evidence showing that subtyping the externalizing disorders, ADHD, ODD and CD, along their comorbidity patterns leads to important differences regarding substance use. This could have implications for the etiology, prevention, and treatment of substance use disorders.

Consommation d'alcool à risque dans un collectif de patients de la Policlinique médicale universitaire de Lausanne [High risk alcohol consumption in a sample of patients of the Lausanne University Medical Polyclinic].

OBJECTIVES AND METHODS: This study indicates the prevalence, the characteristics, and the screening methods of patients with at risk alcohol drinking at the University Medical Clinic of Lausanne. RESULTS: The results reported demonstrate that one patient out of six is a drinker at risk without criteria for alcohol-dependance. The questionnaire AUDIT (Alcohol Use Disorders Identification Test) with a cut-off of five points seems to be the best screening test for at risk alcohol consumption. CONCLUSIONS: The high prevalence of at risk drinking in this study, combined with scientific evidence of the efficiency of brief interventions in changing drinking habits, emphasises the importance of alcohol screening for all patients attending outpatient medical settings.

Pseudomarkets with priorities in large random assignment economies

I study large random assignment economies with a continuum of agents and a finite number of object types. I consider the existence of weak priorities discriminating among agents with respect to their rights concerning the final assignment. The respect for priorities ex ante (ex-ante stability) usually precludes ex-ante envy-freeness. Therefore I define a new concept of fairness, called no unjustified lower chances: priorities with respect to one object type cannot justify different achievable chances regarding another object type. This concept, which applies to the assignment mechanism rather than to the assignment itself, implies ex-ante envy-freeness among agents of the same priority type. I propose a variation of Hylland and Zeckhauser' (1979) pseudomarket that meets ex-ante stability, no unjustified lower chances and ex-ante efficiency among agents of the same priority type. Assuming enough richness in preferences and priorities, the converse is also true: any random assignment with these properties could be achieved through an equilibrium in a pseudomarket with priorities. If priorities are acyclical (the ordering of agents is the same for each object type), this pseudomarket achieves ex-ante efficient random assignments.

Association between circulating cytokine levels, diabetes and insulin resistance in a population-based sample (CoLaus study).

OBJECTIVE: The associations between inflammation, diabetes and insulin resistance remain controversial. Hence, we assessed the associations between diabetes, insulin resistance (using HOMA-IR) and metabolic syndrome with the inflammatory markers high-sensitive C-reactive protein (hs-CRP), interleukin-1 beta (IL-1β), interleukin-6 (IL-6) and tumour necrosis factor-α (TNF-α). DESIGN: Cross-sectional study. PARTICIPANTS: Two thousand eight hundred and eighty-four men and 3201 women, aged 35-75, participated in this study. METHODS: C-reactive protein was assessed by immunoassay and cytokines by multiplexed flow cytometric assay. In a subgroup of 532 participants, an oral glucose tolerance test (OGTT) was performed to screen for impaired glucose tolerance (IGT). RESULTS: IL-6, TNF-α and hs-CRP were significantly and positively correlated with fasting plasma glucose (FPG), insulin and HOMA-IR. Participants with diabetes had higher IL-6, TNF-α and hs-CRP levels than participants without diabetes; this difference persisted for hs-CRP after multivariate adjustment. Participants with metabolic syndrome had increased IL-6, TNF-α and hs-CRP levels; these differences persisted after multivariate adjustment. Participants in the highest quartile of HOMA-IR had increased IL-6, TNF-α and hs-CRP levels; these differences persisted for TNF-α and hs-CRP after multivariate adjustment. No association was found between IL-1β levels and all diabetes and insulin resistance markers studied. Finally, participants with IGT had higher hs-CRP levels than participants with a normal OGTT, but this difference disappeared after controlling for body mass index (BMI). CONCLUSION: We found that subjects with diabetes, metabolic syndrome and increased insulin resistance had increased levels of IL6, TNF-α and hs-CRP, while no association was found with IL-1β. The increased inflammatory state of subjects with IGT is partially explained by increased BMI.