Forecasting

The question of whether the liberalization of the gas industry has led to less concentrated markets has attracted much interest among the scientific community. Classical mathematical regression tools, statistical tests, and optimization equilibrium problems, more precisely non-linear complementarity problems, were used to model European gas markets and their effect on prices. In this research, the parametric and nonparametric game theory methods are employed to study the effect of the market concentration on gas prices. The parametric method takes into account the classical Cournot equilibrium test, with assumptions on cost and demand functions. However, the non-parametric method does not make any prior assumptions, a factor that allows greater freedom in modeling. The results of the parametric method demonstrate that the gas suppliers’ behavior in Austria and The Netherlands gas markets follows the Nash–Cournot equilibrium, where companies act rationally to maximize their payoffs. The non-parametric approach validates the fact that suppliers in both markets follow the same behavior even though one market is more liquid than the other. Interestingly, our findings also suggest that some of the gas suppliers maximize their ‘utility function’ not by only relying on profit, but also on some type of non-profit objective, and possibly collusive behavior. Full article

Many still rightly wonder whether accounting numbers affect business value. Basic questions are “why?” and “how?” We aim at promoting an objective choice on how optimizing the most suitable valuation methods under a “value-based management” framework through some performance measurement systems. First, we present a comprehensive review of valuation methods. Three valuations methods, (i) Free Cash Flow Valuation Model (FCFVM), (ii) Residual Earning Valuation Model (REVM) and (iii) Abnormal Earning Growth Model (AEGM), are presented. We point out advantages and limitations. As applications, the proofs of our findings are illustrated on three study cases: Marks & Spencer’s (M&S’s) business pattern (size and growth prospect), which had a recently advertised valuation “problem”, and two comparable companies, Tesco and Sainsbury’s, all three chosen for multiple-based valuation. For the purpose, two value drivers are chosen, EnV/EBIT (entity value/earnings before interests and taxes) and the corresponding EnV/Sales. Thus, the question whether accounting numbers through models based on mathematical economics truly affect business value has an answer: “Maybe, yes”. Full article

When there exist catchment-wide biases in the distributed hydrologic model states, state updating based on streamflow assimilation at the catchment outlet tends to over- and under-adjust model states close to and away from the outlet, respectively. This is due to the greater sensitivity of the simulated outlet flow to the model states that are located more closely to the outlet in the hydraulic sense, and the subsequent overcompensation of the states in the more influential grid boxes to make up for the larger scale bias. In this work, we describe Mean Field Bias (MFB)-aware variational (VAR) assimilation, or MVAR, to address the above. MVAR performs bi-scale state updating of the distributed hydrologic model using streamflow observations in which MFB in the model states are first corrected at the catchment scale before the resulting states are adjusted at the grid box scale. We comparatively evaluate MVAR with conventional VAR based on streamflow assimilation into the distributed Sacramento Soil Moisture Accounting model for a headwater catchment. Compared to VAR, MVAR adjusts model states at remote cells by larger margins and reduces the Mean Squared Error of streamflow analysis by 2–8% at the outlet Tiff City, and by 1–10% at the interior location Lanagan. Full article

Financial bankruptcy prediction is an essential issue in emerging economies taking into consideration the economic upheaval that can be caused by business failures. The research on bankruptcy prediction is of the utmost importance as it aims to build statistical models that can distinguish healthy firms from financially distressed ones. This paper explores the applicability of the four most used approaches to predict financial bankruptcy using data concerning the case of Greece. A comparison of linear discriminant analysis, logit, decision trees and neural networks is performed. The results show that discriminant analysis is slightly superior to the other methods. Full article

Very-short-term photovoltaic power forecast, namely nowcasting, is gaining increasing attention to face grid stability issues and to optimize microgrid energy management systems in the presence of large penetration of renewable energy sources. In order to identify local phenomena as sharp ramps in photovoltaic production, whole sky images can be used effectively. The first step in the implementation of new and effective nowcasting algorithms is the identification of Sun positions. In this paper, three different techniques (solar angle-based, image processing-based, and neural network-based techniques) are proposed, described, and compared. These techniques are tested on real images obtained with a camera installed at SolarTech^Lab at Politecnico di Milano, Milan, Italy. Finally, the three techniques are compared by introducing some performance parameters aiming to evaluate of their reliability, accuracy, and computational effort. The neural network-based technique obtains the best performance: in fact, this method is able to identify accurately the Sun position and to estimate it when the Sun is covered by clouds. Full article

Considering that most of the photovoltaic (PV) data are behind-the-meter (BTM), there is a great challenge to implement effective demand response projects and make a precise customer baseline (CBL) prediction. To solve the problem, this paper proposes a data-driven PV output power estimation approach using only net load data, temperature data, and solar irradiation data. We first obtain the relationship between delta actual load and delta temperature by calculating the delta net load from matching the net load of irradiation for an approximate day with the least squares method. Then we match and make a difference of the net load with similar electricity consumption behavior to establish the relationship between delta PV output power and delta irradiation. Finally, we get the PV output power and implement PV-load decoupling by modifying the relationship between delta PV and delta irradiation. The case studies verify the effectiveness of the approach and it provides an important reference to perform PV-load decoupling and CBL prediction in a residential distribution network with BTM PV systems. Full article

A semiempirical model, based in the logistic map, was developed to forecast the different phases of the COVID-19 epidemic. This paper shows the mathematical model and a proposal for its calibration. Specific results are shown for Spain. Four phases were considered: non-controlled evolution; total lock-down; partial easing of the lock-down; and a phased lock-down easing. For no control the model predicted the infection of a 25% of the Spanish population, 1 million would need intensive care and 700,000 direct deaths. For total lock-down the model predicted 194,000 symptomatic infected, 85,700 hospitalized, 8600 patients needing an Intensive Care Unit (ICU) and 19,500 deaths. The peak was predicted between the 29 March/3 April. For the third phase, with a daily rate $r=1.03$, the model predicted 400,000 infections and $\mathrm{46,000}\pm \mathrm{15,000}$ deaths. The real r was below 1%, and a revision with updated parameters provided a prediction of 250,000 infected and $\mathrm{29,000}\pm \mathrm{15,000}$ deaths. The reported values by the end of May were 282,870 infected and 28,552 deaths. After easing of the lock-down the model predicted that the health system would not saturate if r was kept below 1.02. This model provided good accuracy during epidemics development. Full article

Traditional learning curve theory assumes a constant learning rate regardless of the number of units produced. However, a collection of theoretical and empirical evidence indicates that learning rates decrease as more units are produced in some cases. These diminishing learning rates cause traditional learning curves to underestimate required resources, potentially resulting in cost overruns. A diminishing learning rate model, namely Boone’s learning curve, was recently developed to model this phenomenon. This research confirms that Boone’s learning curve systematically reduced error in modeling observed learning curves using production data from 169 Department of Defense end-items. However, high amounts of variability in error reduction precluded concluding the degree to which Boone’s learning curve reduced error on average. This research further justifies the necessity of a diminishing learning rate forecasting model and assesses a potential solution to model diminishing learning rates. Full article

The increasing penetration of non-programmable renewable energy sources (RES) is enforcing the need for accurate power production forecasts. In the category of hydroelectric plants, Run of the River (RoR) plants belong to the class of non-programmable RES. Data-driven models are nowadays the most widely adopted methodologies in hydropower forecast. Among all, the Artificial Neural Network (ANN) proved to be highly successful in production forecast. Widely adopted and equally important for hydropower generation forecast is the HYdrological Predictions for the Environment (HYPE), a semi-distributed hydrological Rainfall–Runoff model. A novel hybrid method, providing HYPE sub-basins flow computation as input to an ANN, is here introduced and tested both with and without the adoption of a decomposition approach. In the former case, two ANNs are trained to forecast the trend and the residual of the production, respectively, to be then summed up to the previously extracted seasonality component and get the power forecast. These results have been compared to those obtained from the adoption of a ANN with rainfalls in input, again with and without decomposition approach. The methods have been assessed by forecasting the Run-of-the-River hydroelectric power plant energy for the year 2017. Besides, the forecasts of 15 power plants output have been fairly compared in order to identify the most accurate forecasting technique. The here proposed hybrid method (HYPE and ANN) has shown to be the most accurate in all the considered study cases. Full article

From 2014 to 2018, issuers of retail structured products in Germany established and calculated the Issuer Estimated Value (IEV), a fair value designed to offer more transparency for retail investors. By reporting the IEV in the product information sheet, banks implicitly make a statement on their expected gross margin and, as one of the first papers, we provide an empirical study of the fairness of these disclosed figures. On a sample of discount and capped bonus certificates, we find that reported issuer margins can be verified using standard option pricing models and we illustrate that hedging costs take on an important role for structured product valuation. Consequently, the answer to the raised question in the title seems to be an (initial) ‘yes’ for our chosen product sample. Even though in 2018 the IEV calculations have been replaced by similar margin and cost statements due to the newly introduced Packaged Retail and Insurance-based Investment Products Regulation, this finding might still be a good guide for future research. Full article

Forecasting of forest dynamics at a large scale is essential for land use management, global climate change and biogeochemistry modeling. We develop time series models of the forest dynamics in the conterminous United States based on forest inventory data collected by the US Forest Service over several decades. We fulfilled autoregressive analysis of the basal forest area at the level of US ecological regions. In each USA ecological region, we modeled basal area dynamics on individual forest inventory pots and performed analysis of its yearly averages. The last task involved Bayesian techniques to treat irregular data. In the absolute majority of ecological regions, basal area yearly averages behave as geometric random walk with normal increments. In California Coastal Province, geometric random walk with normal increments adequately describes dynamics of both basal area yearly averages and basal area on individual forest plots. Regarding all the rest of the USA’s ecological regions, basal areas on individual forest patches behave as random walks with heavy tails. The Bayesian approach allowed us to evaluate forest growth rate within each USA ecological region. We have also implemented time series ARIMA models for annual averages basal area in every USA ecological region. The developed models account for stochastic effects of environmental disturbances and allow one to forecast forest dynamics. Full article

Population projections serve various actors at subnational, national, and international levels as a quantitative basis for political and economic decision-making. Usually, the users are no experts in statistics or forecasting and therefore lack the methodological and demographic background to completely understand methods and limitations behind the projections they use to inform further analysis. Our contribution primarily targets that readership. Therefore, we give a brief overview of different approaches to population projection and discuss their respective advantages and disadvantages, alongside practical problems in population data and forecasting. Fundamental differences between deterministic and stochastic approaches are discussed, with special emphasis on the advantages of stochastic approaches. Next to selected projection data available to the public, we show central areas of application of population projections, with an emphasis on Germany. Full article

Hydrological extremes associated with climate change are becoming an increasing concern all over the world. Frequent flooding, one of the extremes, needs to be analyzed while considering climate change to mitigate flood risk. This study forecast streamflow and evaluate risk of flooding in the Neuse River, North Carolina considering future climatic scenarios, and comparing them with an existing Federal Emergency Management Agency study. The cumulative distribution function transformation method was adopted for bias correction to reduce the uncertainty present in the Coupled Model Intercomparison Project Phase 6 (CMIP6) streamflow data. To calculate 100-year and 500-year flood discharges, the Generalized Extreme Value (L-Moment) was utilized on bias-corrected multimodel ensemble data with different climate projections. Out of all projections, shared socio-economic pathways (SSP5-8.5) exhibited the maximum design streamflow, which was routed through a hydraulic model, the Hydrological Engineering Center’s River Analysis System (HEC-RAS), to generate flood inundation and risk maps. The result indicates an increase in flood inundation extent compared to the existing study, depicting a higher flood hazard and risk in the future. This study highlights the importance of forecasting future flood risk and utilizing the projected climate data to obtain essential information to determine effective strategic plans for future floodplain management. Full article

Bus travel time analysis plays a key role in transit operation planning, and methods are needed for investigating its variability and for forecasting need. Nowadays, telematics is opening up new opportunities, given that large datasets can be gathered through automated monitoring, and this topic can be studied in more depth with new experimental evidence. The paper proposes a time-series-based approach for travel time forecasting, and data from automated vehicle monitoring (AVM) of bus lines sharing the road lanes with other traffic in Rome (Italy) and Lviv (Ukraine) are used. The results show the goodness of such an approach for the analysis and reliable forecasts of bus travel times. The similarities and dissimilarities in terms of travel time patterns and city structure were also pointed out, showing the need to take them into account when developing forecasting methods. Full article

We propose to view social conflicts in Africa as having similarities with earthquake occurrences and hence to consider the spatial-temporal Epidemic Type Aftershock Sequence (ETAS) model. The parameters of this highly parameterized model are estimated through simulated annealing. We consider data for 2012 to 2016 to calibrate the model for four African regions separately, and we consider the data for 2017 to evaluate the forecasts. These forecasts concern the amount of future large events as well as their locations. Examples of our findings are that the model predicts a cluster of large events in the Central Africa region, which was not expected based on past events, and that in particular for East Africa it apparently holds that small conflicts can trigger a larger number of conflicts. Full article

Predicting the outcome of sales opportunities is a core part of successful business management. Conventionally, undertaking this prediction has relied mostly on subjective human evaluations in the process of sales decision-making. In this paper, we addressed the problem of forecasting the outcome of Business to Business (B2B) sales by proposing a thorough data-driven Machine-Learning (ML) workflow on a cloud-based computing platform: Microsoft Azure Machine-Learning Service (Azure ML). This workflow consists of two pipelines: (1) An ML pipeline to train probabilistic predictive models on the historical sales opportunities data. In this pipeline, data is enriched with an extensive feature enhancement step and then used to train an ensemble of ML classification models in parallel. (2) A prediction pipeline to use the trained ML model and infer the likelihood of winning new sales opportunities along with calculating optimal decision boundaries. The effectiveness of the proposed workflow was evaluated on a real sales dataset of a major global B2B consulting firm. Our results implied that decision-making based on the ML predictions is more accurate and brings a higher monetary value. Full article

The Integrated Multisatellite Retrievals for Global Precipitation Measurement (GPM) (IMERG) Level 3 estimates rainfall from passive microwave sensors onboard satellites that are associated with several uncertainty sources such as sensor calibration, retrieval errors, and orographic effects. This study aims to provide a comprehensive investigation of multiple machine learning (ML) techniques (Random Forest, and Neural Networks), to stochastically generate an error-corrected improved IMERG precipitation product at a daily time scale and 0.1°-degree spatial resolution over the Brahmaputra river basin. In this study, we used the operational IMERG-Late Run version 06 product along with several meteorological and land surface parameters (elevation, soil type, land type, soil moisture, and daily maximum and minimum temperature) to produce an improved precipitation product in the Brahmaputra basin. We trained, tested, and optimized ML algorithms using 4 years (from 2015 through 2019) of reference rainfall data derived from the rain gauge. The ML generated precipitation product exhibited improved systematic and random error statistics for the study area, which is a strong indication for using the proposed algorithms in retrieving precipitation across the globe. We conclude that the proposed ML-based ensemble framework has the potential to quantify and correct the error sources for improving and promoting the use of satellite-based precipitation estimates for water resources applications. Full article

Improving decision-making in various areas of water policy and management (e.g., flood and drought preparedness, reservoir operation and hydropower generation) requires skillful streamflow forecasts. Despite the recent advances in hydrometeorological prediction, real-time streamflow forecasting over the Himalayas remains a critical issue and challenge, especially with complex basin physiography, shifting weather patterns and sparse and biased in-situ hydrometeorological monitoring data. In this study, we demonstrate the utility of low-complexity data-driven persistence-based approaches for skillful streamflow forecasting in the Himalayan country Nepal. The selected approaches are: (1) simple persistence, (2) streamflow climatology and (3) anomaly persistence. We generated the streamflow forecasts for 65 stream gauge stations across Nepal for short-to-medium range forecast lead times (1 to 12 days). The selected gauge stations were monitored by the Department of Hydrology and Meteorology (DHM) Nepal, and they represent a wide range of basin size, from ~17 to ~54,100 km². We find that the performance of persistence-based forecasting approaches depends highly upon the lead time, flow threshold, basin size and flow regime. Overall, the persistence-based forecast results demonstrate higher forecast skill in snow-fed rivers over intermittent ones, moderate flows over extreme ones and larger basins over smaller ones. The streamflow forecast skill obtained in this study can serve as a benchmark (reference) for the evaluation of many operational forecasting systems over the Himalayas. Full article

This study, which was contracted by the European Commission and is geared towards easy replicability by practitioners, compares the accuracy of individual and combined approaches to forecasting tourism demand for the total European Union. The evaluation of the forecasting accuracies was performed recursively (i.e., based on expanding estimation windows) for eight quarterly periods spanning two years in order to check the stability of the outcomes during a changing macroeconomic environment. The study sample includes Eurostat data from January 2005 until August 2017, and out of sample forecasts were calculated for the last two years for three and six months ahead. The analysis of the out-of-sample forecasts for arrivals and overnights showed that forecast combinations taking the historical forecasting performance of individual approaches such as Autoregressive Integrated Moving Average (ARIMA) models, REGARIMA models with different trend variables, and Error Trend Seasonal (ETS) models into account deliver the best results. Full article