Search

This paper analyses COVID-19 patients’ dynamics during the first wave in the region of Castilla y León (Spain) with around 2.4 million inhabitants using multi-state competing risk survival models. From the date registered as the start of the clinical process, it is assumed that a patient can progress through three intermediate states until reaching an absorbing state of recovery or death. Demographic characteristics, epidemiological factors such as the time of infection and previous vaccinations, clinical history, complications during the course of the disease and drug therapy for hospitalised patients are considered as candidate predictors. Regarding risk factors associated with mortality and severity, consistent results with many other studies have been found, such as older age, being male, and chronic diseases. Specifically, the hospitalisation (death) rate for those over 69 is 27.2% (19.8%) versus 5.3% (0.7%) for those under 70, and for males is 14.5%(7%) versus 8.3%(4.6%)for females. Among patients with chronic diseases the highest rates of hospitalisation are 26.1% for diabetes and 26.3% for kidney disease, while the highest death rate is 21.9% for cerebrovascular disease. Moreover, specific predictors for different transitions are given, and estimates of the probability of recovery and death for each patient are provided by the model. Some interesting results obtained are that for patients infected at the end of the period the hazard of transition from hospitalisation to ICU is significatively lower (p < 0.001) and the hazard of transition from hospitalisation to recovery is higher (p < 0.001). For patients previously vaccinated against pneumococcus the hazard of transition to recovery is higher (p < 0.001). Finally, internal validation and calibration of the model are also performed.

Deep water formation (DWF) in the North Western Mediterranean (NWMed) is a key feature of Mediterranean overturning circulation. Changes in DWF under global warming may have an impact on the region...

Low classroom engagement and distractions are important challenges of massive courses. The literature shows that those problems could decrease when the teacher addresses the students by their names. Learning all the names and faces for a small course is not a difficult task. However, in massive courses, this becomes a tedious and time-consuming task, making it not very practical. This work presents FAIS, a system for effectively learning students names and their corresponding faces, especially designed for massive courses. The system is based on memorization by repetition and gamification techniques, requiring less effort and including a playful perspective. A preliminary experience shows the learning of 105 students names and surnames, just during the daily commutes of the 10 days previous to the start of the course.

Air transportation growth is a reality described by different sources (e.g. The World Bank [1], the latest Eurocontrol report [2]). One essential initiative required to improve air traffic capacity while maintaining or increasing safety is to introduce predictive analytics that enable a dynamic adaptation of airline operations in a preemptive manner to an ever changing environment. An important part of this task is to model airport operations and plan accordingly. Particularly runway usage and/or configuration are important aspects of these operations. For example, prior knowledge of runway usage could improve flight plan optimizers outputs. Of course, to create any model or predictor, ground truth data is required. However most of the time, detailed information about runway historical usage/configuration is inaccessible, unreliable or it belongs to national ATC services providers. Then, thinking on a high-scale forecast methodology there is an important drawback given the lack of a feasible target for most of the airports. Thus, the goal of this work is to introduce an accessible, easy to implement algorithm that allows historical reconstruction of runway usage/configuration for any airport based on data transmitted from aircrafts through either Radar or ADS- B technologies, even when the track data is not consistent. We study the quality of the assessment performed by the two outputs of the algorithm: 1) Measuring runway usage accuracy in comparison to the report given by the Spanish ATC service provider (ENAIRE) for each flight landing to or taking off from two Spanish airports, Madrid-Barajas and Barcelona-El Prat, during October 2016. 2) Comparing the Netherlands-Schiphol runway configuration reported by the Netherlands airspace regulator (LVNL) for three different months: February, April and August 2018. The results provide values above 97% of accuracy for both types of assessment.

A time series clustering algorithm based on the use of the total variation distance between normalized spectra as a measure of dissimilarity is proposed in this work. The oscillatory behavior of the series is thus considered the central characteristic for classication purposes. The proposed algorithm is compared to several other methods which are also based on features extracted from the original series and the results show that its performance is comparable to the best methods available and in some tests it outperforms the rest. As an application the algorithm is used to determine stationary periods for random sea waves, both in simulations and on a real data set, a problem in which changes between stationary sea states are usually slow.

Stochastic ordering among distributions has been considered in a variety of scenarios. However, it is often a restrictive model, not supported by the data even in cases in which the researcher tends to believe that a certain variable is somehow smaller than other. Alternatively, we propose to look at a more flexible version in which two distributions satisfy an approximate stochastic order relation if they are slightly contaminated versions of distributions for which stochastic order holds. The minimal level of contamination required for stochastic order to hold is used as a measure of deviation from exact stochastic order model. Our approach is based on the use of trimmings of probabilities. We discuss their connection to approximate stochastic order and provide theoretical support for its use in data analysis, proving uniform consistency and giving non-asymptotic bounds for the error probabilities of our tests. We provide simulation results and a case study for illustration.

Flight cancellations, departure delays, congestion in taxi times and airborne holding delays are increasingly frequent problems that negatively impact the performance, fuel burn, emissions rate and customer satisfaction at major airports in the world. However, this is just a brushstroke of the future to come. The dramatic growth in the air traffic levels has become a problem of paramount importance, leading into an increased interest for enhancing the current Air Traffic Management (ATM) systems. The main objective is to being able to cope with the sustained air traffic growth under safe, economic, efficient and environmental friendly working conditions. The ADS-B (Automatic Dependent Surveillance — Broadcast) technology plays a major role in the new ATM systems, since it provides more accurate real-time positioning information than secondary radars, in spite of using a cheaper infrastructure. However, the main flaw in the use of ADS-B technology is the generation of large volumes of data, that, when merged with other flight-related information, faces important scalability issues. In this work, we start off from a previously developed data lake for the support of the full ADS-B data life-cycle in a scalable and cost-effective way, and propose a data architecture to integrate data from different providers and reconstruct flight trajectories that can ultimately be used to improve the efficiency in flight operations. This data architecture is also evaluated using a 2-week testbed which reports some interesting figures about its effectiveness.

Comparisons of different treatments or production processes are the goals of a significant fraction of applied research. Unsurprisingly, two sample problems play a main role in statistics through natural questions such as. Is the the new treatment significantly better than the old. However, this is only partially answered by some of the usual statistical tools for this task. More importantly, often practitioners are not aware of the real meaning behind these statistical procedures. We analyze these troubles from the point of view of the order between distributions, the stochastic order, showing evidence of the limitations of the usual approaches, paying special attention to the classical comparison of means under the normal model. We discuss the unfeasibility of statistically proving stochastic dominance, but show that it is possible, instead, to gather statistical evidence to conclude that slightly relaxed versions of stochastic dominance hold. Research partially supported by the Spanish Ministerio de Economía y Competitividad y fondos FEDER, grants MTM2014-56235-C2-1-P and MTM2014-56235-C2-2, and by Consejería de Educación de la Junta de Castilla y León, grant VA212U13.

The grades of a Spanish university access exam involving 10 graders are analyzed. The interest focuses on finding the greatest group of graders showing similar grading patterns or, equivalently, on detecting if there are graders whose grades exhibit significant deviations from the pattern determined by the remaining graders. Due to differences in background of the involved students and graders, homogeneity is too strong to be considered as a realistic null model. Instead, the weaker similarity model, which seems to be more appropriate in this setting, is considered. To handle this problem, a statistical procedure designed to search for a hidden main pattern is developed. The procedure is based on the detection and deletion of the graders that are significantly non-similar to (the pooled mixture of) the others. This is performed through the use of a probability metric, a bootstrap approach and a stepwise search algorithm. Moreover, the procedure also allows one to identify which part of the grades of each grader makes her/him different from the others. © 2013 Elsevier B.V. All rights reserved.

The use of trimming procedures constitutes a natural approach to robustifying statistical methods. This is the case of goodness-of-fit tests based on a distance, which can be modified by choosing trimmed versions of the distributions minimizing that distance. The L"2-Wasserstein distance is used to introduce the trimming methodology for assessing when a data sample can be considered mostly normal. The method can be extended to other location and scale models, introducing a robust approach to model validation, and allows an additional descriptive analysis by determining the subset of the data with the best improved fit to the model. This is a consequence of the use of data-driven trimming methods instead of the more classical symmetric trimming procedures.

In this paper, we introduce logistic models to analyse fertility curves. The models are formulated as linear models of the log odds of fertility and are defined in terms of parameters that are interpreted as measures of level, location and shape of the fertility schedule. This parameterization is useful for the evaluation, and interpretation of fertility trends and projections of future period fertility. For a series of years, the proposed models admit a state-space formulation that allows a coherent joint estimation of parameters and forecasting. The main features of the models compared with other alternatives are the functional simplicity, the flexibility, and the interpretability of the parameters. These and other features are analysed in this paper using examples and theoretical results. Data from different countries are analysed, and to validate the logistic approach, we compare the goodness of fit of the new model against well-known alternatives; the analysis gives superior results in most developed...

This article introduces an analysis of similarity of distributions based on the L2-Wasserstein distance between trimmed distributions. Our main innovation is the use of the impartial trimming methodology, already considered in robust statistics, which we adapt to this setup. Instead of simply removing data at the tails to provide some robustness to the similarity analysis, we develop a data-driven trimming method aimed at maximizing similarity between distributions. Dissimilarity is then measured in terms of the distance between the optimally trimmed distributions. We provide illustrative examples showing the improvements over previous approaches and give the relevant asymptotic results to justify the use of this methodology in applications.

An algorithm for determining stationary periods for time series of random sea waves is proposed in this work. This is a problem in which changes between stationary sea states are usually slow and segmentation procedures based on change-point detection frequently give poor results. The method is based on a new procedure for time series clustering, built on the use of the total variation distance between normalized spectra as a measure of dissimilarity. The oscillatory behavior of the series is thus considered the central characteristic for classification purposes. The proposed algorithm is compared to several other methods which are also based on features extracted from the original series and the results show that its performance is comparable to the best methods available and in some tests it performs better. This clustering method may be of independent interest.

Statistical methods of dimension reduction and classification are used to obtain homogeneous local-area clustering with regard to the most relevant demographic parameters. The dimension reduction is conducted in two stages using Principal Component Analysis and a modified k-mean procedure is proposed to determine the final clusters. This clustering will be useful in future demographic studies at a local level, in particular to obtain forecasts of demographic rates and population projections. The region of Castile and Leon in Spain is used to illustrate the method. A Poisson model is used to explore the advantages of the new clustering over the more conventional classification based on provinces.

The problem of detecting changes in the state of the sea is very important for the analysis and determination of wave climate in a given location. Wave measurements are frequently statistically analyzed as a time series, and segmentation algorithms developed in this context are used to determine change-points. However, most methods found in the literature consider the case of instantaneous changes in the time series, which is not usually the case for sea waves, where changes take a certain time interval to occur. We propose a new segmentation method that allows for the presence of transition intervals between successive stationary periods, and is based on the analysis of distances of normalized spectra to detect clusters in the time series. The series is divided into 30-minutes intervals and the spectral density is estimated for each one. The normalized spectra are compared using the Total Variation distance and a hierarchical clustering method is applied to the distance matrix. The information obtained from the clustering algorithm is used to classify the intervals as belonging to a stationary or a transition period We present simulation studies to validate the method and examples of applications to real data.

Stochastic ordering among distributions has been considered in a variety of scenarios. Economic studies often involve research about the ordering of investment strategies or social welfare. However, as noted in the literature, stochastic orderings are often a too strong assumption which is not supported by the data even in cases in which the researcher tends to believe that a certain variable is somehow smaller than other. Instead of considering this rigid model of stochastic order we propose to look at a more flexible version in which two distributions are said to satisfy an approximate stochastic order relation if they are slightly contaminated versions of distributions which do satisfy the stochastic ordering. The minimal level of contamination that makes this approximate model hold can be used as a measure of the deviation of the original distributions from the exact stochastic order model. Our approach is based on the use of trimmings of probability measures. We discuss the connection between them and the approximate stochastic order model and provide theoretical support for its use in data analysis. We also provide simulation results.

Quality of life (QOL) and quality of work life (QOWL) play a key role in the overall concept of sustainability. In this paper we analyze QOL and QOWL variables in relation to the type of livestock farm (organic/conventional) and the use of some quality label (PDO Protected Designation of Origin, PGI Protected Geographical Indication). Data were collected through regular visits in 2011 to 70 small ruminant farms in Spain using five-point Likert items with 1-5 range. Of all the 70 farms, 6 are organics, 39 have a quality label and 17 use traditional manufacturing methods. The possession of quality labels in the small ruminant farms analyzed seems not to be related with the quality of life and work. Farmers with traditional production show higher valuation of quality of life but not of quality of work. Organic farms provide quality of life and quality of work significantly better than conventional ones.

We say that two probabilities are similar at level $\alpha$ if they are contaminated versions (up to an $\alpha$ fraction) of the same common probability. We show how this model is related to minimal distances between sets of trimmed probabilities. Empirical versions turn out to present an overfitting effect in the sense that trimming beyond the similarity level results in trimmed samples that are closer than expected to each other. We show how this can be combined with a bootstrap approach to assess similarity from two data samples., Comment: Published in at http://dx.doi.org/10.3150/11-BEJ351 the Bernoulli (http://isi.cbs.nl/bernoulli/) by the International Statistical Institute/Bernoulli Society (http://isi.cbs.nl/BS/bshome.htm)

Tesis (Ingeniería Comercial) Expresar en cifras la presencia, comportamiento e importancia de las pequeñas y medianas empresas (PYME) en mercado de bienes y servicios chileno, requiere de una definición previa acerca del criterio con el cual se acotará este segmento empresarial. Existen diversas opciones para determinar los tamaños de empresas: empleo, volumen de ventas, capital y otros; incluso se pueden construir clasificadores que combinen más de una variable. En este trabajo se ha seleccionado el criterio de clasificación utilizado en el libro sobre la PYME publicado por CORFO en 1994. . Es decir, una clasificación de las empresas según el volumen de ventas anuales que realizan. Las razones que llevaron en aquella publicación a elegir esa variable continúan siendo válidas. Este indicador entrega una medida bastante cercana del nivel de actividad desarrollado por cada empresa y de su potencial acceso al sistema financiero, a la tecnología y a la capacitación. Además, esa información sigue siendo la única disponible que hace referencia al universo de empresas del país. Fijado el volumen de ventas como criterio de clasificación, las subdivisiones por estratos se efectuaron según la definición de tamaño de empresas que CORFO emplea en la actualidad en la mayoría de sus programas de apoyo a la PYME. Estimaciones realizadas por la CORFO, indican que a comienzos de los años noventa, las exportaciones realizadas por la PYME, representaban alrededor del 4% del total de los bienes exportados por el país. A partir de 1994, considerando que fue modificada la definición de las pequeñas y medianas empresas los datos no son comparables con los años anteriores. La participación de las pequeñas empresas se incrementa de 2,04% en 1990 hasta 2,4% en 1993, para reducirse posteriormente hasta llegar al 1,3% en 1997. Este resultado es consecuencia del hecho de que los montos exportados por estas unidades productivas, después de haber crecido en valores corrientes hasta 1993 se quedaron estancados (hasta se redujeron), mientras las exportaciones totales del país casi llegaron a duplicarse.