Forecasting, Volume 4, Issue 2 (June 2022) – 9 articles

Compared to other industries, fashion apparel retail faces many challenges in predicting future demand for its products with a high degree of precision. Fashion products’ short life cycle, insufficient historical information, highly uncertain market demand, and periodic seasonal trends necessitate the use of models that can contribute to the efficient forecasting of products’ sales and demand. Many researchers have tried to address this problem using conventional forecasting models that predict future demands using historical sales information. While these models predict product demand with fair to moderate accuracy based on previously sold stock, they cannot fully be used for predicting future demands due to the transient behaviour of the fashion industry. This paper proposes an intelligent forecasting system that combines image feature attributes of clothes along with its sales data to predict future demands. The data used for this empirical study is from a European fashion retailer, and it mainly contains sales information on apparel items and their images. The proposed forecast model is built using machine learning and deep learning techniques, which extract essential features of the product images. The model predicts weekly sales of new fashion apparel by finding its best match in the clusters of products that we created using machine learning clustering based on products’ sales profiles and image similarity. The results demonstrated that the performance of our proposed forecast model on the tested or test items is promising, and this model could be effectively used to solve forecasting problems. Full article

Because of the non-linearity inherent in energy commodity prices, traditional mono-scale smoothing methodologies cannot accommodate their unique properties. From this viewpoint, we propose an extended mode decomposition method useful for the time-frequency analysis, which can adapt to various non-stationarity signals relevant for enhancing forecasting performance in the era of big data. To this extent, we employ variants of mode decomposition-based extreme learning machines namely: (i) Complete Ensemble Empirical Mode Decomposition with Adaptive Noise-based ELM Model (CEEMDAN-ELM), (ii) Ensemble Empirical Mode Decomposition-based ELM Model (EEMD-ELM) and (iii) Empirical Mode Decomposition Based ELM Model (EMD-ELM), which cut-across soft computing and artificial intelligence to analyze multi-commodity time series data by decomposing them into seven independent intrinsic modes and one residual with varying frequencies that depict some interesting characterization of price volatility. Our findings show that in terms of the model-specific forecast accuracy measures different dynamics in the two scenarios namely the (non) COVID periods. However, the introduction of a benchmark, namely the autoregressive integrated moving average model (ARIMA) reveals a slight change in the earlier dynamics, where ARIMA outperform our proposed models in the Japan gas and the US gas markets. To check the superiority of our models, we apply the model-confidence set (MCS) and the Kolmogorov-Smirnov Predictive Ability test (KSPA) with more preference for the former in a multi-commodity framework, which reveals that in the pre-COVID era, CEEMDAN-ELM shows persistence and superiority in accurately forecasting Crude oil, Japan gas, and US gas. Nonetheless, this paradigm changed during the COVID-era, where CEEMDAN-ELM favored Japan gas, US gas, and coal market with different rankings via the Model confidence set evaluation methods. Overall, our numerical experiment indicates that all decomposition-based extreme learning machines are superior to the benchmark model. Full article

The path choice models play a key role in transportation engineering, especially when coupled with an assignment procedure allowing link flows to be obtained. Their implementation could be complex and resource-consuming. In particular, such a task consists of several stages, including (1) the collection of a large set of data from surveys to infer users’ path choices and (2) the definition of a model able to reproduce users’ choice behaviors. Nowadays, stage (1) can be improved using floating car data (FCD), which allow one to obtain a reliable dataset of paths. In relation to stage (2), different structures of models are available; however, a compromise has to be found between the model’s ability to reproduce the observed paths (including the ability to forecast the future path choices) and its applicability in real contexts (in addition to guaranteeing the robustness of the assignment procedure). Therefore, the aim of this paper is to explore the opportunities offered by FCD to calibrate a path/route choice model to be included in a general procedure for scenario assessment. The proposed methodology is applied to passenger and freight transport case studies. Significant results are obtained showing the opportunities offered by FCD in supporting path choice simulation. Moreover, the characteristics of the model make it easily applicable and exportable to other contexts. Full article

Forecast developers predominantly assess residuals and error statistics when tuning the targeted model’s quality. With that, eventual cost or rewards of the underlying business application are typically not considered in the model development phase. The analysis of the power system wholesale market allows us to translate a time series forecast method’s quality to its respective business value. For instance, near real-time capacity procurement takes place in the wholesale market, which is subject to complex interrelations of system operators’ grid activities and balancing parties’ scheduling behavior. Such forecasting tasks can hardly be solved with model-driven approaches because of the large solution space and non-convexity of the optimization problem. Thus, we generate load forecasts by means of a data-driven based forecasting tool ProLoaF, which we benchmark with state-of-the-art baseline models and the auto-machine learning models auto.arima and Facebook Prophet. Full article

Companies facing transformation in the automotive industry will need to adapt to new trends, technologies and functions, in order to remain competitive. The challenge is to anticipate such trends and to forecast their development over time. The aim of this paper is to develop a methodology that allows us to analyze the temporal development of technologies, taking automated driving as an example. The framework consists of a technological and a functional roadmap. The technology roadmap provides information on the temporal development of 59 technologies based on expert elicitation using a multi-stage Delphi survey and patent analyses. The functional roadmap is derived from a meta-analysis of studies including 209 predictions of the maturity of automated driving functions. The technological and functional roadmaps are merged into a consolidated roadmap, linking the temporal development of technologies and functions. Based on the publication analysis, SAE level 5 is predicted to be market-ready by 2030. Contrasted to the results from the Delphi survey in the technological roadmap, 2030 seems to be too optimistic, however, as some key technologies would not have reached market readiness by this time. As with all forecasts, the proposed framework is not able to accurately predict the future. However, the combination of different forecast approaches enables users to have a more holistic view of future developments than with single forecasting methods. Full article

The paper applies innovation diffusion models to study the adoption process of solar PV energy in the UK from 2010 to 2021 by comparing the trajectories between three main categories, residential, commercial, and utility, in terms of both the number of installations and installed capacity data. The effect of the UK incentives on adoptions by those categories is studied by analyzing the timing, intensity, and persistence of the perturbations on adoption curves. The analysis confirms previous findings on PV adoption, namely the fragile role of the media support to solar PV, the ability of the proposed model to capture both the general trend of adoptions and the effects induced by ad hoc incentives, and the dramatic dependence of solar PV from public incentives. Thanks to the granularity of the data, the results reveal several interesting aspects, related both to differences in adoption patterns depending on the category considered, and to some regularities across categories. A comparison between the models for number of installations and for installed capacity data suggests that the latter (usually more easily available than the former) may be highly informative and, in some cases, may provide a reliable description of true adoption data. Full article

This paper proposes the application of a multivariate diffusion model, based on ordinary differential equations, to investigate the energy transition in Germany. Specifically, the model is able to analyze the dynamic interdependencies between coal, gas and renewables in the energy market. A system dynamics representation of the model is also performed, allowing a deeper understanding of the system and the set-up of suitable strategic interventions through a simulation exercise. Such simulation provides a useful indication of how renewable energy consumption may be stimulated as a result of well-specified policies. Full article

Tourism is one of the main sources of wealth for the Moroccan regions, since, in 2019, it contributed 7.1% to the total GDP. However, it is considered to be one of the sectors most vulnerable to exogenous shocks (political and social stability, currency change, natural disasters, pandemics, etc.). To control this, policymakers tend to use various techniques to forecast tourism demand for making crucial decisions. In this study, we aimed to forecast the number of tourist arrivals to the Marrakech-Safi region using annual data for the period from 1999 to 2018 by using three conventional approaches (ARIMA, AR, and linear regression), and then we compared the results with three artificial intelligence-based techniques (SVR, XGBoost, and LSTM). Then, we developed hybrid models by combining both the conventional and AI-based models, using the technique of ensemble learning. The findings indicated that the hybrid models outperformed both conventional and AI-based techniques. It is clear from the results that using hybrid models can overcome the limitations of each method individually. Full article

Since its inception in 2009, Bitcoin has increasingly gained main stream attention from the general population to institutional investors. Several models, from GARCH type to jump-diffusion type, have been developed to dynamically capture the price movement of this highly volatile asset. While fitting the Gaussian and the Generalized Hyperbolic and the Normal Inverse Gaussian (NIG) distributions to log-returns of Bitcoin, NIG distribution appears to provide the best fit. The time-varying Hurst parameter for Bitcoin price reveals periods of randomness and mean-reverting type of behaviour, motivating the study in this paper through fractional Ornstein–Uhlenbeck driven by a Normal Inverse Gaussian Lévy process. Features such as long-range memory are jump diffusion processes that are well captured with this model. The results present a 95% prediction for the price of Bitcoin for some specific dates. This study contributes to the literature of Bitcoin price forecasts that are useful for Bitcoin options traders. Full article