David Conesa

Sensitivity analysis of efficiency and Malmquist productivity indices : An application to Spanish savings banks

Hypothesis testing and statistical precision in the context of nonparametric efficiency and productivity measurement have been investigated since the early 1990s. Recent contributions focus on this matter through the use of resampling methods?i.e., bootstrapping techniques. However, empirical evidence is still practically non-existent. This gap is more noticeable in the case of banking efficiency studies, where the literature is immense. In this paper, we explore productivity growth and productive efficiency for Spanish savings banks over the (initial) post-deregulation period 1992?1998 using Data Envelopment Analysis (DEA) and bootstrapping techniques. Results show that productivity growth has occurred, mainly due to improvement in production possibilities, and that mean efficiency has remained fairly constant over time. The bootstrap analysis yields further evidence, as for many firms productivity growth, or decline, is not statistically significant. With regard to efficiency measurement, the bootstrap reveals that the disparities in the original efficiency scores of some firms are lessened to a great extent. Desde principios de los anos noventa ha habido avances notables en el contraste de hipotesis dentro del contexto de la medicion de la eficiencia y la productividad mediante tecnicas parametricas. Las contribuciones mas recientes han enfocado el tema a traves de metodos de remuestreo -conocidos en la literatura como tecnicas bootstrap-. Sin embargo, practicamente no ha habido aplicaciones, algo tambien patente en el estudio de la eficiencia de la empresa bancaria. En este articulo, analizamos la eficiencia productiva y el crecimiento de la productividad de las cajas de ahorro espanolas durante el periodo 1992-1998 a traves de tecnicas no parametricas (DEA) y de tecnicas bootstrap, con el fin de poder realizar inferencia estadistica. Los resultados indican que la productividad ha aumentado, principalmente debido a una mejora en las posibilidades de produccion, mientras que la eficiencia promedio no ha variado sustancialmente. El analisis bootstrap revela que, en el caso de la productividad, para muchas empresas su aumento o disminucion no es estadisticamente significativo. En cuanto a la eficiencia, muestra que las diferencias entre empresas individuales se reducen de manera notable cuando consideramos intervalos de confianza.

Bayesian Markov switching models for the early detection of influenza epidemics

The early detection of outbreaks of diseases is one of the most challenging objectives of epidemiological surveillance systems. In this paper, a Markov switching model is introduced to determine the epidemic and non-epidemic periods from influenza surveillance data: the process of differenced incidence rates is modelled either with a first-order autoregressive process or with a Gaussian white-noise process depending on whether the system is in an epidemic or in a non-epidemic phase. The transition between phases of the disease is modelled as a Markovian process. Bayesian inference is carried out on the former model to detect influenza epidemics at the very moment of their onset. Moreover, the proposal provides the probability of being in an epidemic state at any given moment. In order to validate the methodology, a comparison of its performance with other alternatives has been made using influenza illness data obtained from the Sanitary Sentinel Network of the Comunitat Valenciana, one of the 17 autonomous regions in Spain.

Bovine paramphistomosis in Galicia (Spain): Prevalence, intensity, aetiology and geospatial distribution of the infection

The present study explored various basic aspects of the epidemiology of paramphistomosis in Galicia, the main cattle producing region in Spain. In total, 589 cows from different farms located across the region were selected at random in the slaughterhouse for examination of the rumens and reticula for the presence of Paramphistomidae flukes. Paramphistomes were found in 111 of 589 necropsied cows (18.8%; 95% CI: 15.7-21.9%), with higher prevalences of infection in beef cows than in dairy cows (29.2% vs 13.9%). Although the number of flukes per animal was generally low (median=266 flukes), some cows harboured large parasite burdens (up to 11,895 flukes), which may have harmful effects on their health or productivity. Cows with higher parasite burdens also excreted greater numbers of fluke eggs in their faeces, which suggests that heavily parasitized mature cows play an important role in the transmission of paramphistomosis. This role may be particularly important in Galicia, where the roe deer, which is the only wild ruminant in the study area, was found not to be a reservoir for the infection. The use of morpho-anatomical and molecular techniques applied to a large number of fluke specimens provided reliable confirmation that Calicophoron daubneyi is the only species of the family Paramphistomidae that parasitizes cattle in Galicia. The environmental data from the farms of origin of the necropsied cows were used in Bayesian geostatistical models to predict the probability of infection by C. daubneyi throughout the region. The results revealed the role of environmental risk factors in explaining the geographical heterogeneity in the probability of infection in beef and dairy cattle. These explanatory factors were used to construct predictive maps showing the areas with the highest predicted risk of infection as well as the uncertainty associated with the predictions.

Estimation and prediction of the spatial occurrence of fish species using Bayesian latent Gaussian models

A methodological approach for modelling the occurrence patterns of species for the purpose of fisheries management is proposed here. The presence/absence of the species is modelled with a hierarchical Bayesian spatial model using the geographical and environmental characteristics of each fishing location. Maps of predicted probabilities of presence are generated using Bayesian kriging. Bayesian inference on the parameters and prediction of presence/absence in new locations (Bayesian kriging) are made by considering the model as a latent Gaussian model, which allows the use of the integrated nested Laplace approximation (INLA) software (which has been seen to be quite a bit faster than the well-known MCMC methods). In particular, the spatial effect has been implemented with the stochastic partial differential equation (SPDE) approach. The methodology is evaluated on Mediterranean horse mackerel (Trachurus mediterraneus) in the Western Mediterranean. The analysis shows that environmental and geographical factors can play an important role in directing local distribution and variability in the occurrence of species. Although this approach is used to recognize the habitat of mackerel, it could also be for other different species and life stages in order to improve knowledge of fish populations and communities.

Prediction in Markovian bulk arrival queues

This paper deals with the statistical analysis of bulk arrival queues from a Bayesian point of view. The focus is on prediction of the usual measures of performance of the system in equilibrium. Posterior predictive distribution of the number of customers in the system is obtained through its probability generating function. Posterior distribution of the waiting time, in the queue and in the system, of the first customer of an arriving group is expressed in terms of their Laplace and Laplace–Stieltjes transform. Discussion of numerical inversion of these transforms is addressed.

On the dynamics of eco-efficiency performance in the European Union

This paper evaluates the evolution of environmental performance in the context of the European Union (EU), over the period 1993-2010. The context is particularly relevant, due to the traditionally high concerns of the EU about these issues, which has triggered off several initiatives and regulations on environmental protection. In this setting, we conduct a two-stage analysis which develops environmental performance indicators in the first stage for each pair country-year, and evaluates its evolution in the second. More specifically, in the first stage we estimate specific efficiencies for three air-pollutants (CO2e, SO2, NOx), along with an eco-efficiency indicator, for which we use the slack-free directional distance functions in the Data Envelopment Analysis framework (as opposed to the more extended intensity ratios), whereas in the second stage we propose to using a model of explicit distribution dynamics which takes into account how the entire distributions of these indicators evolve. Our results indicate that the dynamics underlying the evolution of the indicators analyzed are indeed remarkable. Although the eco-efficiency indicator has improved over the last two decades, it has been during the last decade when performance has shown a more convergent path. However, in the case of the more traditional indicators (CO2e, SO2, NOx) the abatement opportunities are still remarkable, especially in the case of SO2. HighlightsWe evaluate the evolution of environmental performance in the context of the EU.We estimate specific efficiencies for CO2e, SO2, NOx and an eco-efficiency indicator.We propose using a model of explicit distribution dynamics.Results indicate that the dynamics of the indicators are complex.Despite the recent progresses, the abatement opportunities are still remarkable.

Bootstrapping profit change: An application to Spanish banks

The aim of this study is to provide a tool which enables us to conduct statistical analysis in the context of changes in productivity and profit. We build on previous initiatives to decompose profit change into mutually exclusive and exhaustive sources. To do this we use distance functions, which are calculated empirically using linear programming techniques. However, we may not learn a great deal by solving these linear programs unless methods of statistical analysis are used to examine the properties of the relevant estimators. Our purpose is to provide a methodology based on bootstrap that allows us to conduct statistical inference for the profit change decomposition. Thus, it will be possible to answer questions such as whether variations in the profit change components, or the differences across firms, are statistically significant. We provide an application to Spanish commercial banks for the 2003/2004 period. Results suggest that profit change differentials between them are not always significant. Therefore, the validity of the conclusions which do not factor in the bootstrap may be jeopardized to varying degrees.

Statistical performance of a multiclass bulk production queueing system

Abstract In this paper, we discuss how to statistically analyze a make-to-stock production system the behaviour of which depends on a multiclass bulk queueing system. The performance of the system is evaluated in terms of the different demands of products, processing times and, mainly, through the finished product inventory and other related measures that quantify the queueing effects in the system. A numerical example which illustrates the applicability of the results in an inventory scenario is also discussed.

Predicting the behaviour of the renal transplant waiting list in the País Valencià (Spain) using simulation modeling

A discrete event simulation model has been set up in order to analyze the renal transplant waiting list in the Pais Valencia, one of the autonomous regions in which Spain is divided. The model combines the information of the arrival of the patients onto the list and the process of donations, which also depend on the number of kidneys provided by each donor. Bayesian inference has been used to take into account the uncertainty about the parameters of the input distributions (acceptance, donation and transplantation rates). After validating the model, predictions about the future behaviour of the waiting list have been done. Results indicate a decrease in the size of the waiting list in a short and middle term. Comparison with other strategies of simulation has been done in order to confirm the problem of underestimation of the variance of the expected simulation output.

Bayesian hierarchical Poisson models with a hidden Markov structure for the detection of influenza epidemic outbreaks

Considerable effort has been devoted to the development of statistical algorithms for the automated monitoring of influenza surveillance data. In this article, we introduce a framework of models for the early detection of the onset of an influenza epidemic which is applicable to different kinds of surveillance data. In particular, the process of the observed cases is modelled via a Bayesian Hierarchical Poisson model in which the intensity parameter is a function of the incidence rate. The key point is to consider this incidence rate as a normal distribution in which both parameters (mean and variance) are modelled differently, depending on whether the system is in an epidemic or non-epidemic phase. To do so, we propose a hidden Markov model in which the transition between both phases is modelled as a function of the epidemic state of the previous week. Different options for modelling the rates are described, including the option of modelling the mean at each phase as autoregressive processes of order 0, 1 or 2. Bayesian inference is carried out to provide the probability of being in an epidemic state at any given moment. The methodology is applied to various influenza data sets. The results indicate that our methods outperform previous approaches in terms of sensitivity, specificity and timeliness.