Predictability of drug expenditures: An application using morbidity data

The growth of pharmaceutical expenditure and its prediction is a major concern for policy makers and health care managers. This paper explores different predictive models to estimate future drug expenses, using demographic and morbidity individual information from an integrated healthcare delivery organization in Catalonia for years 2002 and 2003. The morbidity information consists of codified health encounters grouped through the Clinical Risk Groups (CRGs). We estimate pharmaceutical costs using several model specifications, and CRGs as risk adjusters, providing an alternative way of obtaining high predictive power comparable to other estimations of drug expenditures in the literature. These results have clear implications for the use of risk adjustment and CRGs in setting the premiums for pharmaceutical benefits.

Bridging DSGE models and the raw data

A method to estimate DSGE models using the raw data is proposed. The approachlinks the observables to the model counterparts via a flexible specification which doesnot require the model-based component to be solely located at business cycle frequencies,allows the non model-based component to take various time series patterns, andpermits model misspecification. Applying standard data transformations induce biasesin structural estimates and distortions in the policy conclusions. The proposed approachrecovers important model-based features in selected experimental designs. Twowidely discussed issues are used to illustrate its practical use.

Crash Data Validation: An Iowa Case Study, February, 2007

With the quickening pace of crash reporting, the statistical editing of data on a weekly basis, and the ability to provide working databases to users at CTRE/Iowa Traffic Safety Data Service, the University of Iowa, and the Iowa DOT, databases that would be considered incomplete by past standards of static data files are in “public use” even as the dynamic nature of the central DOT database allows changes to be made to both the aggregate of data and to the individual crashes already reported. Moreover, “definitive” analyses of serious crashes will, by their nature, lag seriously behind the preliminary data files. Even after these analyses, the dynamic nature of the mainframe data file means that crash numbers can continue to change long after the incident year. The Iowa DOT, its Office of Driver Services (the “data owner”), and institutional data users/distributors must establish data use, distribution, and labeling protocols to deal with the new, dynamic nature of data. In order to set these protocols, data must be collected concerning the magnitude of difference between database records and crash narratives and diagrams. This study determines the difference between database records and crash narratives for the Iowa Department of Transportation’s Office of Traffic and Safety crash database and the impacts of this difference.

Mapping of Environmental Data Using Kernel-Based Methods

Recently, kernel-based Machine Learning methods have gained great popularity in many data analysis and data mining fields: pattern recognition, biocomputing, speech and vision, engineering, remote sensing etc. The paper describes the use of kernel methods to approach the processing of large datasets from environmental monitoring networks. Several typical problems of the environmental sciences and their solutions provided by kernel-based methods are considered: classification of categorical data (soil type classification), mapping of environmental and pollution continuous information (pollution of soil by radionuclides), mapping with auxiliary information (climatic data from Aral Sea region). The promising developments, such as automatic emergency hot spot detection and monitoring network optimization are discussed as well.

Status of Young Women in the Juvenile Justice System Preliminary Data, 2007

Young women involved in the juvenile justice system present with characteristics and experiences that differentiate them from their male counterparts. As such, the juvenile justice system in Iowa must consider these factors if it is to effectively and efficiently impact recidivism and rehabilitation.

Evaluating predictive densities of U.S. output growth and inflation in a large macroeconomic data set

We evaluate conditional predictive densities for U.S. output growth and inflationusing a number of commonly used forecasting models that rely on a large number ofmacroeconomic predictors. More specifically, we evaluate how well conditional predictive densities based on the commonly used normality assumption fit actual realizationsout-of-sample. Our focus on predictive densities acknowledges the possibility that, although some predictors can improve or deteriorate point forecasts, they might have theopposite effect on higher moments. We find that normality is rejected for most modelsin some dimension according to at least one of the tests we use. Interestingly, however,combinations of predictive densities appear to be correctly approximated by a normaldensity: the simple, equal average when predicting output growth and Bayesian modelaverage when predicting inflation.

Does direct democracy reduce the size of government? New evidence from historical data, 1890-2000

Using historical data for all Swiss cantons from 1890 to 2000, we estimate the causal effect of direct democracy on government spending. The main innovation in this paper is that we use fixed effects to control for unobserved heterogeneity and instrumental variables to address the potential endogeneity of institutions. We find that the budget referendum and lower costs to launch a voter initiative are effective tools in reducing canton level spending. However, we find no evidence that the budget referendum results in more decentralized government or a larger local government. Our instrumental variable estimates suggest that a mandatory budget referendum reduces the size of canton spending between 13 and 19 percent. A 1 percent lower signature requirement for the initiative reduces canton spending by up to 2 percent.

Geographical effects on the accuracy of textile trade data:an international approach for 1913

Foreign trade statistics are the main data source to the study of international trade.However its accuracy has been under suspicion since Morgernstern published hisfamous work in 1963. Federico and Tena (1991) have resumed the question arguing thatthey can be useful in an adequate level of aggregation. But the geographical assignmentproblem remains unsolved. This article focuses on the spatial variable through theanalysis of the reliability of textile international data for 1913. A geographical biasarises between export and import series, but because of its quantitative importance it canbe negligible in an international scale.

An analysis of the accounting principles applied by the European Farm Accountancy Data Network

In spite of its relative importance in the economy of many countriesand its growing interrelationships with other sectors, agriculture has traditionally been excluded from accounting standards. Nevertheless, to support its Common Agricultural Policy, for years the European Commission has been making an effort to obtain standardized information on the financial performance and condition of farms. Through the Farm Accountancy Data Network (FADN), every year data are gathered from a rotating sample of 60.000 professional farms across all member states. FADN data collection is not structured as an accounting cycle but as an extensive questionnaire. This questionnaire refers to assets, liabilities, revenues and expenses, and seems to try to obtain a "true and fair view" of the financial performance and condition of the farms it surveys. However, the definitions used in the questionnaire and the way data is aggregated often appear flawed from an accounting perspective. The objective of this paper is to contrast the accounting principles implicit in the FADN questionnaire with generally accepted accounting principles, particularly those found in the IVth Directive of the European Union, on the one hand, and those recently proposed by the International Accounting Standards Committee’s Steering Committeeon Agriculture in its Draft Statement of Principles, on the other hand. There are two reasons why this is useful. First, it allows to make suggestions how the information provided by FADN could be more in accordance with the accepted accounting framework, and become a more valuable tool for policy makers, farmers, and other stakeholders. Second, it helps assessing the suitability of FADN to become the starting point for a European accounting standard on agriculture.

Determinants of economic growth: Will data tell?

Many factors inhibiting and facilitating economic growth havebeen suggested. Can agnostics rely on international incomedata to tell them which matter? We find that agnostic priorslead to conclusions that are sensitive to differences acrossavailable income estimates. For example, the PWT 6.2 revisionof the 1960-96 income estimates in the PWT 6.1 leads tosubstantial changes regarding the role of government,international trade, demography, and geography. We concludethat margins of error in international income estimates appeartoo large for agnostic growth empirics.

Fracture network characterization using hydrological and geophysical data

Les décisions de gestion des eaux souterraines doivent souvent être justiffées par des modèles quantitatifs d'aquifères qui tiennent compte de l'hétérogénéité des propriétés hydrauliques. Les aquifères fracturés sont parmi les plus hétérogènes et très difficiles à étudier. Dans ceux-ci, les fractures connectées, d'ouverture millimètrique, peuvent agir comme conducteurs hydrauliques et donc créer des écoulements très localisés. Le manque général d'informations sur la distribution spatiale des fractures limite la possibilité de construire des modèles quantitatifs de flux et de transport. Les données qui conditionnent les modèles sont généralement spatialement limitées, bruitées et elles ne représentent que des mesures indirectes de propriétés physiques. Ces limitations aux données peuvent être en partie surmontées en combinant différents types de données, telles que les données hydrologiques et de radar à pénétration de sol plus commun ément appelé géoradar. L'utilisation du géoradar en forage est un outil prometteur pour identiffer les fractures individuelles jusqu'à quelques dizaines de mètres dans la formation. Dans cette thèse, je développe des approches pour combiner le géoradar avec les données hydrologiques affn d'améliorer la caractérisation des aquifères fracturés. Des investigations hydrologiques intensives ont déjà été réalisées à partir de trois forage adjacents dans un aquifère cristallin en Bretagne (France). Néanmoins, la dimension des fractures et la géométrie 3-D des fractures conductives restaient mal connue. Affn d'améliorer la caractérisation du réseau de fractures je propose dans un premier temps un traitement géoradar avancé qui permet l'imagerie des fractures individuellement. Les résultats montrent que les fractures perméables précédemment identiffées dans les forages peuvent être caractérisées géométriquement loin du forage et que les fractures qui ne croisent pas les forages peuvent aussi être identiffées. Les résultats d'une deuxième étude montrent que les données géoradar peuvent suivre le transport d'un traceur salin. Ainsi, les fractures qui font partie du réseau conductif et connecté qui dominent l'écoulement et le transport local sont identiffées. C'est la première fois que le transport d'un traceur salin a pu être imagé sur une dizaines de mètres dans des fractures individuelles. Une troisième étude conffrme ces résultats par des expériences répétées et des essais de traçage supplémentaires dans différentes parties du réseau local. En outre, la combinaison des données de surveillance hydrologique et géoradar fournit la preuve que les variations temporelles d'amplitude des signaux géoradar peuvent nous informer sur les changements relatifs de concentrations de traceurs dans la formation. Par conséquent, les données géoradar et hydrologiques sont complémentaires. Je propose ensuite une approche d'inversion stochastique pour générer des modèles 3-D de fractures discrètes qui sont conditionnés à toutes les données disponibles en respectant leurs incertitudes. La génération stochastique des modèles conditionnés par géoradar est capable de reproduire les connexions hydrauliques observées et leur contribution aux écoulements. L'ensemble des modèles conditionnés fournit des estimations quantitatives des dimensions et de l'organisation spatiale des fractures hydrauliquement importantes. Cette thèse montre clairement que l'imagerie géoradar est un outil utile pour caractériser les fractures. La combinaison de mesures géoradar avec des données hydrologiques permet de conditionner avec succès le réseau de fractures et de fournir des modèles quantitatifs. Les approches présentées peuvent être appliquées dans d'autres types de formations rocheuses fracturées où la roche est électriquement résistive.

PyRate: a new program to estimate speciation and extinction rates from incomplete fossil data

Despite the advancement of phylogenetic methods to estimate speciation and extinction rates, their power can be limited under variable rates, in particular for clades with high extinction rates and small number of extant species. Fossil data can provide a powerful alternative source of information to investigate diversification processes. Here, we present PyRate, a computer program to estimate speciation and extinction rates and their temporal dynamics from fossil occurrence data. The rates are inferred in a Bayesian framework and are comparable to those estimated from phylogenetic trees. We describe how PyRate can be used to explore different models of diversification. In addition to the diversification rates, it provides estimates of the parameters of the preservation process (fossilization and sampling) and the times of speciation and extinction of each species in the data set. Moreover, we develop a new birth-death model to correlate the variation of speciation/extinction rates with changes of a continuous trait. Finally, we demonstrate the use of Bayes factors for model selection and show how the posterior estimates of a PyRate analysis can be used to generate calibration densities for Bayesian molecular clock analysis. PyRate is an open-source command-line Python program available at http://sourceforge.net/projects/pyrate/.