On some pre-test and Stein-rule phi-divergence test estimators in the independence model of categorical data

For the model of independence in a two way contingency table, shrinkage estimators based on minimum φ$\phi$-divergence estimators and φ$\phi$-divergence statistics are considered. These estimators are based on the James–Stein-type rule and incorporate the idea of preliminary test estimator. The asymptotic bias and risk are obtained under contiguous alternative hypotheses, and on the basis of them a comparison study is carried out.     

Shortcomings in the Measurement of Innovation: Implications for Accounting Standard Setting

Innovation is nowadays a fundamental determinant of value creation in business companies and economic growth. Therefore, the measurement of innovation has become a significant concern both for business companies and governments. Traditionally, attempts to measure innovation have adopted a macroeconomic approach, as they have been largely based on broad surveys. However, no attempt has been made to date in order to complement the information provided by such surveys with aggregated data obtained from the financial reports of individual companies. This paper analyses the conceptual and methodological problems underlying the measurement of business innovation by means of surveys and discusses the lack of ability of accounting standards to accurately reflect innovative activities in the financial statements of business firms.In the light of the evidence provided by the empirical studies published to date, we analyse the Spanish situation by reviewing innovation studies conducted by the National Institute of Statistics (INE), and assessing the relationship between the value relevance of accounting information and the firm's technological level. Our results suggest that both, micro- and macroeconomic approaches towards the measurement of innovation have significant shortcomings. Thus, a joint effort seems to be needed in order to overcome the methodological limitations affecting innovation studies based on surveys and those relying on financial accounting information. Despite their limitations, surveys provide a sound basis for the identification of trends, key factors and explanatory variables. On the other hand, financial statements could provide a sound basis for the measurement of innovation if they included more relevant information on the intangible determinants of the value of companies. This has obvious implications for the standard setting process.     

Preliminary phi-divergence test estimators for linear restrictions in a logistic regression model

The problem of estimation of the parameters in a logistic regression model is considered under multicollinearity situation when it is suspected that the parameter of the logistic regression model may be restricted to a subspace. We study the properties of the preliminary test based on the minimum ϕ -divergence estimator as well as in the ϕ -divergence test statistic. The minimum ϕ -divergence estimator is a natural extension of the maximum likelihood estimator and the ϕ -divergence test statistics is a family of the test statistics for testing the hypothesis that the regression coefficients may be restricted to a subspace.     

Minimum phi-divergence estimators for loglinear models with linear constraints and multinomial sampling

In this paper the family ofφ-divergence estimators for loglinear models with linear constraints and multinomial sampling is studied. This family is an extension of the maximum likelihood estimator studied by Haber and Brown (1986). A simulation study is presented and some alternative estimators to the maximum likelihood are obtained. This work was parcially supported by Grant DGES PB2003-892     

Lean-healthcare approach to reduce costs in a sterilization plant based on surgical tray rationalization

Abstract

Sterilization plants process all reusable materials used in hospitals during different stages of healthcare provision. Most items sterilized come from the surgical theater in the form of individual instruments, linen and surgical trays. Reducing the number of instruments in trays impacts on the performance of the sterilization process. Here, we propose a three-phase method, based on lean healthcare principles, cluster analysis and kaizen groups of experts to rationalize surgical trays, reducing sterilization processing costs. We innovate by proposing a strategy to approach tray rationalization prioritizing surgical specialties that are both complex and frequent, and by using group technology to cluster surgical trays expediting expert analyses carried out by kaizen groups. Our propositions were tested in a high complexity, tertiary care hospital, covering 12 surgical specialties, reducing the number of instruments by average 9.75%, and the time-to-assemble trays by average 9.68%, obtaining annual savings of US$285.756,00 in sterilization processing costs.     

Ergonomics at Volkswagen Brasil. Multidisciplinary work to equalize health, productivity and quality

Following the technological developments and presented to the multidisciplinary processes as automakers, Volkswagen Brazil, represented by its ergonomists, through this paper, shows the importance of ergonomic efficiency of management applied to various levels of life of a product, since its creation to its final implementation on production lines.The preventive work of ergonomists during the processes leading to production of a vehicle is accomplished through assessments on a simulated system called the Digital Factory. Since the initial stages to final product delivery there is a need for a multidisciplinary aligning the concepts of ergonomics, productivity and quality of product. Industrial Engineering, Process Engineering, ergonomists and workers are involved in the analysis made through the Workshop's showing the importance of discussion between the various users of the systems. The processes of series are also equipped with a set of certifications flow of job and planned audits on items that describe processes and applied ergonomics.     

Residual analysis and outliers in loglinear models based on phi-divergence statistics

In this paper we consider new families of residuals and influential measures, under the assumption of multinomial sampling, for loglinear models. These new families are based on φ$\phi$-divergence test statistic. The asymptotic normality of the standardized residuals is obtained as well as the relation of the new family of influential measures with the appropriate Cook's distance in this context. The expression of the new family of residuals is obtained in two important problems: independence and symmetry in two-dimensional contingence tables. A numerical example illustrates the results obtained.     

Fertility and nuptiality changes in Spain from the late 18th to the early 20th century

Abstract 3.1. A regional approach often reveals features of population trends not evident in national data. We have already pointed out that the 1768 census followed the ecclesiastical sub-divisions of the country; therefore its territorial data are not comparable with those derived from later enumerations. The 1787 and 1797 censuses, on the other hand, were based on civil sub-divisions, which can be compared, when aggregated, with later censuses of the modern statistical era.