Abstract
Performance estimation aims at estimating the loss that a predictive model will incur on unseen data. This process is a fundamental stage in any machine learning project. In this paper we study the application of these methods to time series forecasting tasks. For independent and identically distributed data the most common approach is cross-validation. However, the dependency among observations in time series raises some caveats about the most appropriate way to estimate performance in this type of data. Currently, there is no consensual approach. We contribute to the literature by presenting an extensive empirical study which compares different performance estimation methods for time series forecasting tasks. These methods include variants of cross-validation, out-of-sample (holdout), and prequential approaches. Two case studies are analysed: One with 174 real-world time series and another with three synthetic time series. Results show noticeable differences in the performance estimation methods in the two scenarios. In particular, empirical experiments suggest that blocked cross-validation can be applied to stationary time series. However, when the time series are non-stationary, the most accurate estimates are produced by out-of-sample methods, particularly the holdout approach repeated in multiple testing periods.

Abstract
In this article, a multilevel single-phase converter is proposed and investigated. Its structure is based on cascaded-transformer systems, which are very interesting in applications in which a single dc source is available. The features of well-known multilevel cascaded H-bridge and transformer-based solutions are integrated into the proposed converter. As a result, the number of synthesized voltage levels can be optimized without excessively increasing the number of transformers. The basic configuration has six two-level insulated-gate bipolar transistor legs, two injection transformers, and two dc links (the lowest voltage dc link may be a floating capacitor or be connected to a small dc power source). The configuration is generalized and the calculation of the transformer's turns ratios as well as the dc-link voltages to maximize the number of voltage levels is provided. The proposed configuration is compared with cascaded H-bridge and a single-phase shared leg converter, which are also cascaded by means of transformers. Compared with the conventional converters, the proposed one has lower switching losses and higher conduction losses. Thus, the proposed configuration is more interesting in terms of semiconductor losses for high-voltage and low-current applications. Experimental and simulation results are shown to demonstrate the feasibility of the system.

Abstract
Deterministic Synchronous Multichannel Extension (DSME) is a prominent MAC behavior first introduced in IEEE 802.15.4e supporting deterministic guarantees using its multisuperframe structure. DSME also facilitates techniques like multi-channel and Contention Access Period (CAP) reduction to increase the number of available guaranteed timeslots in a network. However, any tuning of these functionalities in dynamic scenarios is not explored in the standard. In this paper, we present a multisuperframe tuning technique called DynaMO which tunes the CAP reduction and Multisuperframe Order in an effective manner to improve flexibility and scalability, while guaranteeing bounded delay. We also provide simulations to prove that DynaMO with its dynamic tuning feature can offer up to 15--30% reduction in terms of latency in a large DSME network.

Abstract
Within the framework of the entrepreneurial university (EU), this study aims to test its multidimensional domain and therefore to confirm the positive contribution of EU factors to perceived regional competitiveness in Portugal. Data were collected from ten Portuguese public universities (PPUs) through a self-employed questionnaire. First- and second-order confirmatory factor analysis (CFA) were performed through factor and multiple linear regression analyses. The proposed EU construct was confirmed, thus proving the adequacy of scales for the PPUs context. Overall, the main findings show that EU factors—‘internal processes’, ‘entrepreneurial supporting measures’, ‘international collaboration’, and ‘funding strategy’—make a positive contribution to the perception of regional competitiveness. ‘Entrepreneurial supporting measures’ is the EU factor which has the biggest impact on perceived regional competitiveness and ‘organisational design’ is the only EU factor that does not reveal any impact on it. This contribution demonstrates to policy makers that PPUs are not merely cost centres but knowledge spillovers that can have a positive influence on regional competitiveness. © Springer Nature Switzerland AG 2020.

Abstract

Abstract
This article presents an original predictive strategy, based on a new mid-term forecasting model, to be used for trading physical electricity futures. The forecasting model is used to predict the average spot price, which is used to estimate the Risk Premium corresponding to electricity futures trade operations with a physical delivery. A feed-forward neural network trained with the extreme learning machine algorithm is used as the initial implementation of the forecasting model. The predictive strategy and the forecasting model only need information available from electricity derivatives and spot markets at the time of negotiation. In this paper, the predictive trading strategy has been applied successfully to the Iberian Electricity Market (MIBEL). The forecasting model was applied for the six types of maturities available for monthly futures in the MIBEL, from 1 to 6 months ahead. The forecasting model was trained with MIBEL price data corresponding to 44 months and the performances of the forecasting model and of the predictive strategy were tested with data corresponding to a further 12 months. Furthermore, a simpler forecasting model and three benchmark trading strategies are also presented and evaluated using the Risk Premium in the testing period, for comparative purposes. The results prove the advantages of the predictive strategy, even using the simpler forecasting model, which showed improvements over the conventional benchmark trading strategy, evincing an interesting hedging potential for electricity futures trading.