Search

Today, the higher education sector can be considered a market and, within it, private university education is a common marketable service in the literature on higher education management. Research on the analysis of the variables that generate the university image has been the subject of numerous investigations. Although there is no generally accepted definition, most authors approach the measurement of image through multi-factor scales, with variables relating to functional and psychological elements. This research aims to contribute to study of the most determinant variables in measuring a product’s image, assessing especially the effect of the reputation construct. This was done through measuring the image of the “private university” product as perceived by citizens of Andalusia, based on a standardized model with three dimensions—functional and affective aspects and reputation. After adapting and validating the questionnaire, a two-phase procedure is performed with double validation through exploratory and confirmatory factor analysis. The results show an adapted scale valid for measuring the image of a generic product; with presentation and discussion of a series of advantages of incorporating reputation and measuring image through models with three dimensions. This article goes deeper into the possible influence of reputation as a determinant factor in measuring image, an assumption arising from some models for measuring image, something that so far has not been sufficiently contrasted.

This paper examines the out-of-sample performance of thirteen portfolio strategies in the Spanish financial market (Ibex 35) during the COVID-19 pandemic (considering four different periods) by usi...

Even though the evolution of regional economic activity is substantially explained by national/common economic shocks, regional economies can still engage in significant differential or divergent b...

Extreme learning machine (ELM) has shown to be a suitable algorithm for classification problems. Several ensemble meta-algorithms have been developed in order to generalize the results of ELM models. Ensemble approaches introduced in the ELM literature mainly come from boosting and bagging frameworks. The generalization of these methods relies on data sampling procedures, under the assumption that training data are heterogeneously enough to set up diverse base learners. The proposed ELM ensemble model overcomes this strong assumption by using the negative correlation learning (NCL) framework. An alternative diversity metric based on the orthogonality of the outputs is proposed. The error function formulation allows us to develop an analytical solution to the parameters of the ELM base learners, which significantly reduce the computational burden of the standard NCL ensemble method. The proposed ensemble method has been validated by an experimental study with a variety of benchmark datasets, comparing it with the existing ensemble methods in ELM. Finally, the proposed method statistically outperforms the comparison ensemble methods in accuracy, also reporting a competitive computational burden (specially if compared to the baseline NCL-inspired method).

The traditional machine-part cell formation problem simultaneously clusters machines and parts in different production cells from a zero–one incidence matrix that describes the existing interactions between the elements. This manuscript explores a novel alternative for the well-known machine-part cell formation problem in which the incidence matrix is composed of non-binary values. The model is presented as multiple-ratio fractional programming with binary variables in quadratic terms. A simple reformulation is also implemented in the manuscript to express the model as a mixed-integer linear programming optimization problem. The performance of the proposed model is shown through two types of empirical experiments. In the first group of experiments, the model is tested with a set of randomized matrices, and its performance is compared to the one obtained with a standard greedy algorithm. These experiments showed that the proposed model achieves higher fitness values in all matrices considered than the greedy algorithm. In the second type of experiment, the optimization model is evaluated with a real-world problem belonging to Human Resource Management. The results obtained were in line with previous findings described in the literature about the case study.

COVID-19 has had a negative impact on the living conditions of people in all countries worldwide. With a devastating economic crisis where many families are finding it difficult to pay bills and make ends meet, increases in prices of food basket staples can be very worrying. This study examines the relationship between the incidence of the pandemic during the first wave in 16 Eurozone countries with the variation experienced in food prices. We analysed the harmonised index of consumer food prices (included in HICP) and the classification of the degree of pandemic impact by country, the latter established with the index of deaths provided by the Johns Hopkins Center. The procedure used compared actual food prices during the first wave (March to June 2020) with those foreseeable in the absence of the pandemic. Time series analysis was used, dividing the research period into two phases. In both phases, the Holt–Winters model was applied for estimation and subsequent prediction. After a contrast using Kendall’s tau correlation index, it was concluded that in the countries with the highest death rates during the first wave, there was a higher increase in food prices than in the least affected countries of the Eurozone.

This work analyses the complementarity and contrast between two metrics commonly used for evaluating the quality of a binary classifier: the correct classification rate or accuracy, C, and the F1 metric, which is very popular when dealing with imbalanced datasets. Based on this analysis, a set of constraints relating C and F1 are defined as a function of the ratio of positive patterns in the dataset. We evaluate the possibility of using a multi-objective evolutionary algorithm guided by this pair of metrics to optimise binary classification models. To check the validity of the constraints, we perform an empirical analysis considering 26 benchmark datasets obtained from the UCI repository and an interesting liver transplant dataset. The results show that the relation is fulfilled and that the use of the algorithm for simultaneously optimising the pair (C,F1) leads to a generally balanced accuracy for both classes. The experiments also reveal that, in some cases, better results are obtained by using the majority class as the positive class instead of using the minority one, which is the most common approach with imbalanced datasets.

Increasing operational efficiency is an objective relevant for all institutions, but it is essential in public entities and even more in public health systems because of the number of resources they consume and their impact on general welfare. This research analyses the effect that CEOs’ gender has on the operational efficiency of the entities they manage. Despite the impact that the management team and notably the CEO have on the development of institutions, studies on their effect on performance are practically non-existent, especially for public organisations. We have used data from acute care hospital trusts belonging to the English National Health System (NHS) concerning its development. The results were obtained from a two-stage analysis. First, the entities’ economic efficiency and health/social efficiency (two operational efficiency measures) were evaluated using two data envelopment analysis (DEA) models. Secondly, the results have been regressed with the CEOs’ gender. The results obtained are robust and consistent, revealing that male CEOs have greater performance than female CEOs. This result provides insight into determining features that relate to operational efficiency, which it is of interest to the research and policymakers

Ensembles are a widely implemented approach in the machine learning community and their success is traditionally attributed to the diversity within the ensemble. Most of these approaches foster diversity in the ensemble by data sampling or by modifying the structure of the constituent models. Despite this, there is a family of ensemble models in which diversity is explicitly promoted in the error function of the individuals. The negative correlation learning (NCL) ensemble framework is probably the most well-known algorithm within this group of methods. This article analyzes NCL and reveals that the framework actually minimizes the combination of errors of the individuals of the ensemble instead of minimizing the residuals of the final ensemble. We propose a novel ensemble framework, named global negative correlation learning (GNCL), which focuses on the optimization of the global ensemble instead of the individual fitness of its components. An analytical solution for the parameters of base regressors based on the NCL framework and the global error function proposed is also provided under the assumption of fixed basis functions (although the general framework could also be instantiated for neural networks with nonfixed basis functions). The proposed ensemble framework is evaluated by extensive experiments with regression and classification data sets. Comparisons with other state-of-the-art ensemble methods confirm that GNCL yields the best overall performance.

In this paper, we have developed a classification model for online learning environments that relates the Instructors Overall Performance (IOP) rating (according to students’ perceptions) with the course characteristics, students’ demographics and the effectiveness of the instructor in his/her teaching roles. To that end, a comprehensive Student Evaluation of Teaching (SET) instrument is proposed, which includes not only conventional teaching elements, but also items that encourage twenty-first century skills. The goal of the study is twofold: (i) to quantify the extent to which the selected variables explain the IOP rating, and (ii) determine which teaching and non-teaching variables most affect the IOP rating. The best performing classifier achieved a competitive accuracy, highlighting that the selected variables mainly determine the IOP values. Other important findings include: (i) the IOP value is mainly influenced by the effectiveness of the instructor in his/her teaching roles; (ii) teaching...

Accuracy has been used traditionally to evaluate the performance of classifiers. However, it is well known that accuracy is not able to capture all the different factors that characterize the performance of a multiclass classifier. In this manuscript, accuracy is studied and analyzed as a weighted average of the classification rate of each class. This perspective allows us to propose the dispersion of the classification rate of each class as its complementary measure. In this sense, a graphical performance metric, which is defined in a two dimensional space composed by accuracy and dispersion, is proposed to evaluate the performance of classifiers. We show that the combined values of accuracy and dispersion must fall within a clearly bounded two dimensional region, different for each problem. The nature of this region depends only on the a priori probability of each class, and not on the classifier used. Thus, the performance of multiclassifiers is represented in a two dimensional space where the models can be compared in a more fair manner, providing greater awareness of the strategies that are more accurate when trying to improve the performance of a classifier. Furthermore we experimentally analyze the behavior of seven different performance metrics based on the computation of the confusion matrix values in several scenarios, identifying clusters and relationships between measures. As shown in the experimentation, the graphical metric proposed is specially suitable in challenging, highly imbalanced and with a high number of classes datasets. The approach proposed is a novel point of view to address the evaluation of multiclassifiers and it is an alternative to other evaluation measures used in machine learning.

Nowadays, a lot of studies are trying to measure happiness. There are many indexes that measure progress. We have chosen two of them: Happy Planet Index (HPI) and World Happiness Report (WHR). The first of them, an index from New Economics Foundation, is a measure of sustainable well-being in terms of how countries support their citizens to live good lives now and in the future. It uses global data on experienced well-being, life expectancy, and Ecological Footprint from 151 countries. On the other hand, the WHR, an index from the United Nations and Columbia University for 156 countries, uses six factors to explain this: Gross Domestic Product (GDP) per capita, years of healthy life expectancy, social support, perceptions of corruption, generosity, and freedom to make life choices. Both indexes have used different methodology to measure happiness so some countries present different levels of happiness depending on the index that is used. The first objective of this chapter is to propose a new index, using factor analysis to classify 145 common countries between HPI and WHR and to simplify the ranking in four parts. We found that there are areas like South America that could be considered “happy,” independently of the index (HPI, WHR, new index) and vice versa. Therefore, because there is not a unified measure of happiness at the moment we try to reach an agreement between both of them with this new index. Without a doubt, worrying about happiness, at an international level, has caused many countries to fight for a privileged place in the rankings published periodically. Big corporations and global brands are also aware of this situations, where, very often, an association is made between brand’s names and attributes like subjective well-being or happiness; a struggle exist in the consumer’s mind because brands want to be perceived as vectors to reach the subjective and/or collective well-being. Thus, the second objective of this chapter is to confirm that there exist an association between the level of happiness of a country, according to the new index proposed, and the valuation that consumers make of brands’ contribution to their level of well-being, taking the last Havas’s group study Meaningful brands. This work also delves in the relation between the optimism percentage about the future situation and the valuation that citizens do of brands contribution in their well-being level.

The mean-variance (MV) portfolio is typically formulated as a quadratic programming (QP) problem that linearly combines the conflicting objectives of minimizing the risk and maximizing the expected return through a risk aversion profile parameter. In this formulation, the two objectives are expressed in different units, an issue that could definitely hamper obtaining a more competitive set of portfolio weights. For example, a modification in the scale in which returns are expressed (by one or percent) in the MV portfolio, implies a modification in the solution of the problem. Motivated by this issue, a novel mean squared variance (MSV) portfolio is proposed in this paper. The associated optimization problem of the proposed strategy is very similar to the Markowitz optimization, with the exception of the portfolio mean, which is presented in squared form in our formulation. The resulting portfolio model is a non-convex QP problem, which has been reformulated as a mixed-integer linear programming (MILP) problem. The reformulation of the initial non-convex QP problem into an MILP allows for future researchers and practitioners to obtain the global solution of the problem via the use of current state-of-the-art MILP solvers. Additionally, a novel purely data-driven method for determining the optimal value of the hyper-parameter that is associated with the MV and MSV approaches is also proposed in this paper. The MSV portfolio has been empirically tested on eight portfolio time series problems with three different estimation windows (composing a total of 24 datasets), showing very competitive performance in most of the problems.

In this paper, the neural network version of Extreme Learning Machine (ELM) is used as a base learner for an ensemble meta-algorithm which promotes diversity explicitly in the ELM loss function. The cost function proposed encourages orthogonality (scalar product) in the parameter space. Other ensemble-based meta-algorithms from AdaBoost family are used for comparison purposes. Both accuracy and diversity presented in our proposal are competitive, thus reinforcing the idea of introducing diversity explicitly.