Inaccurate Forecasts Research Articles

Quantifying Time-Varying Forecast Uncertainty and Risk for the Real Price of Oil

We propose a novel and numerically efficient quantification approach to forecast uncertainty of the real price of oil using a combination of probabilistic individual model forecasts. Our combination method extends earlier approaches that have been applied to oil price forecasting, by allowing for sequentially updating of time-varying combination weights, estimation of time-varying forecast biases and facets of miscalibration of individual forecast densities and time-varying inter-dependencies among models. To illustrate the usefulness of the method, we present an extensive set of empirical results about time-varying forecast uncertainty and risk for the real price of oil over the period 1974–2018. We show that the combination approach systematically outperforms commonly used benchmark models and combination approaches, both in terms of point and density forecasts. The dynamic patterns of the estimated individual model weights are highly time-varying, reflecting a large time variation in the relative performance of the various individual models. The combination approach has built-in diagnostic information measures about forecast inaccuracy and/or model set incompleteness, which provide clear signals of model incompleteness during three crisis periods. To highlight that our approach also can be useful for policy analysis, we present a basic analysis of profit-loss and hedging against price risk.

Short-Term Electricity Load Forecasting Based on Temporal Fusion Transformer Model

Electricity load forecasting plays an important role in the operation of power systems. Inaccurate forecast would reduce the safety of power supply and affect the economic and social activities as well as national defense and security. In addition, the forecast results also support decision-making on electricity generation and market transactions. Traditional methods such as AR, ARIMA, SARIMA have been widely used to forecast short term electricity load. Recently, load forecasting based on artificial and deep neural networks have shown significant accuracy improvement over traditional statistical models. In this research, a novel recurrent neural network named temporal fusion transformer (TFT) is used to forecast short-term electricity load of Hanoi city. The TFT is a newly developed model and it combines the advantages of several other RNN models such as LSTM and the self-attention mechanism. In addition to historical load data, we use temperature and humidity features, and time features such as calendar month, lunar month, days of the week, hours of the day and holidays. The forecast results of TFT are compared with traditional statistical models as well as well-known RNN models. The compared results show that the proposed method is better than other methods in both MAE and MAPE criteria.

Supply Chain Demand Forecast Based on SSA-XGBoost Model

Supply chain management usually faces problems such as high empty rate of transportation, unreasonable inventory management, and large material consumption caused by inaccurate market demand forecasts. To solve these problems, using artificial intelligence and big data technology to achieve market demand forecasting and intelligent decision-making is becoming a strategic technology trend of supply chain management in the future. Firstly, this paper makes a visual analysis of the historical data of the Stock Keeping Unit (SKU); Then, the characteristic factors affecting the future demand are constructed from the storage level, product level, historical usage of SKU, etc; Finally, a supply chain demand forecasting algorithm based on SSA-XGBoost model has proposed around three aspects of feature engineering, parameter optimization and model integration, and is compared with other machine learning models. The experiment shows that the forecasting result of SSA-XGBoost forecasting model is highly consistent with the actual value, so it is of practical significance to adopt this forecasting model to solve the supply chain demand forecasting problem.

Toward Improving the Efficiency of Software Development Effort Estimation via Clustering Analysis

<i>Introduction</i>: The precise estimation of software effort is a significant difficulty that project managers encounter during software development. Inaccurate forecasting leads to either overestimating or underestimating software effort, which can be detrimental for stakeholders. The International Function Point Users Group&#x2019;s Function Point Analysis (FPA) method is one of the most critical methods for software effort estimation. However, the practice of using the FPA method in the same fashion across all software areas needs to be reexamined. <i>Aim</i>: We propose a model for evaluating the influence of data clustering on software development effort estimation and then finding the best clustering method. We call this model the effort estimation using machine learning applied to the clusters (EEAC) model. <i>Method</i>: We cluster the dataset according to the clustering method and then apply the FPA and EEAC methods to these clusters for effort estimation. The clustering methods we use in this study include five categorical variable criteria (Development Platform, Industrial Sector, Language Type, Organization Type, and Relative Size) and the k-means clustering algorithm. <i>Results</i>: The experimental results show that the estimation accuracy obtaining with clustering consistently outperforms the accuracy without clustering for both the FPA and EEAC methods. Significantly, using the FPA method, the average improvement rate from using clustering as opposed to non-clustered was highest at 58.06%, according to the RMSE. With the EEAC method, this number reached 65.53%. The Industry Sector categorical variable achieves the best accuracy estimation compared to the other clustering criteria and k-means clustering. The improvement in accuracy in terms of the RMSE when applying this criterion is 63.68% for the FPA method and 72.02% for the EEAC method. <i>Conclusion</i>: Better results are obtained through dataset clustering compared to no clustering for both the FPA and EEAC methods. The Industry Sector is the most suitable clustering method among the tested clustering methods.

Machine learning-based cost predictive model for better operating expenditure estimations of U.S. light rail transit projects

Inaccurate forecasts of operating expenditures (OPEX) during the planning phase for new Light Rail Transit (LRT) projects in the United States underestimated future costs by up to 45% ( Harmatuck, 2008 ). When OPEX exceeded projected levels, local transit agencies often reduced public transit services to operate within their respective annual budgets. Therefore, it is imperative for transit agencies to produce reasonably accurate planning estimates to secure sufficient funding to support future operations, maintenance, and service delivery associated with LRT systems. The research aimed to develop a more accurate LRT OPEX predictive model to be used during the planning stage. Traditional statistical analysis and various machine learning-based algorithms were utilized with input from 22 LRT systems in the United States spanning between 2008 and 2018 from various U.S. governmental public databases. This study extended the current state of practice that relied primarily on sum of unit-cost estimates (also known as the unit-cost method) which generally failed to produce accurate forecasts due to lack of engineering details at the planning stage. Existing research attempted to develop regression-based methodologies using system-based attributes but did not substantially increase prediction accuracy from using the unit-cost method. The research improved current practices and research by having developed a more accurate and replicable machine learning-based predictive model using available geographic, socio-economic and LRT system-related variables at the planning stage.

ARTIFICIAL NEURAL NETWORK BASED DEMAND FORECASTING INTEGRATED WITH FEDERAL FUNDS RATE

Adverse effects of inaccurate demand forecasts; stockouts, overstocks, customer loss have led academia and the business world towards accurate demand forecasting methods. Artificial Neural Network (ANN) is capable of highly accurate forecasts integrated with many variables. The use of Price and Promotion variables have increased the accuracy while the addition of other relevant variables would decrease the occurrences of errors. The use of the Federal Funds Rate as an additional macroeconomic variable to ANN forecasting models has been discussed in this research by the means of the accuracy measuring method: Average Relative Mean Absolute Error.

CONCEPTUAL PRINCIPLES OF FISCAL EFFICIENCY OF INDIRECT TAXATION IN UKRAINE

The article proposes the concept of fiscal efficiency of indirect taxation, the purpose of which is to ensure growth and stabilization of fiscal significance of indirect taxes in consolidated budget revenues and GDP of Ukraine, growth of VAT fiscal efficiency indicators (productivity, effective rate, VAT budget efficiency ratio). taxes. The concept is based on the general and specific principles of indirect taxation, defines the functions, purpose, objectives and methods aimed at assessing and identifying reserves of fiscal efficiency of indirect taxation. The indicators of fiscal significance of indirect taxes in the dynamics are assessed, the level of their influence on the formation of the state budget revenues and the country’s GDP is determined. It is determined that the highest level of fiscal significance among indirect taxes in VAT, followed by excise tax and duty. The analysis revealed a slight reduction in the fiscal importance of indirect taxes, due to the crisis in the country. The structure of VAT revenues to the state budget of Ukraine is presented, the factors that influenced the non-fulfillment of planned VAT revenues are identified. Thus, inaccurate forecast of inflation and exchange rates, reduced imports of goods, reduced sales of tobacco products, the effect of anti-crisis measures on Covid-19, reduced consumer demand, closing borders, the situation on the world and national markets and more. The method of calculation is given and the indicators of fiscal efficiency of VAT are estimated, the downward trend of the indicator of productivity of VAT, effective rate and coefficient of budgetary efficiency of VAT is defined. It is determined that VAT revenues have the potential for growth, as there is a significant gap between the effective and regulatory VAT rate.

Analysis of PCA based AdaBoost Machine Learning Model for Predict Mid-Term Weather Forecasting

In general, weather forecasting is done with the use of enormously complicated physical science models that use a variety of environmental circumstances over a long period of time. Because of the annoyances of the climatic framework, these criteria are frequently fragile, causing the models to produce inaccurate forecasts. The models are mostly run on multiple hubs in a massive High-Performance Computing (HPC) environment that uses a lot of energy. In this research, we offer a climate expectation approach that uses historical data from various climate stations to create basic AI models that may provide meaningful forecasts for specific climatic conditions in the not-too-distant future within a given time frame. In this paper, we offer a climate expectation approach that uses historical data from several climate stations to create basic AI models that can anticipate certain climatic conditions in the not-too-distant future within a given time frame. Overall research performed into two stages; in first stage Principle component Analysis has been used to extract the irrelevant attributes from the datasets. In second stage five different machines learning algorithm used to predict temperature condition for midterm span &amp; finally four performance indicators along with training time used to identify the best fitted model. From the result analysis it is seen that PCA based AdaBoost model is the fittest model with acquired the best outcome of R2, RMSE, MAE &amp; MSE are 0.992, 0.539, 0.398 &amp; 0.209 respectively. Beside of this present model also outperformed than the other state of art model proposed for midterm weather forecasting purpose. Keyword : Weather Forecasting, PCA, Machine learning, Performance indicator

Experimental testing of a real aggregator system performing rigorous optimal control of electrical and thermal storage

This paper provides an experimentally validated optimal control approach based on a Hamilton–Jacobi–Bellman (HJB) model for optimising aggregated distributed energy resources across multiple energy carriers. The research incorporates nonlinear effects arising from storage degradation, conversion efficiency and self-discharge as well as multiple energy carrier storage. A semi-Lagrangian HJB solver was implemented on low-cost digital controllers, and integrated into a real fully functional cloud-based aggregation platform. The computational cost is kept at a minimum, enabling on-line computations on the low-cost controller, while maintaining a rigorous proof of convergence to the theoretical value function of the nonlinear, non-convex optimal control problem.The controller links into a distributed optimal control platform that is using local as well as cloud-based information and performs all the computation and decision-making locally. The distributed controllers were tested and validated on site with an electrical and a thermal storage device. Experimental results confirm that the framework is practical, accommodates nonlinear effects and inaccurate external forecasts, has a small computational cost, is robust and can deliver significant cost benefits to the stakeholders.

Centralized - Decentralized Demand Management And The Bullwhip Effect

Information distortion tends to inflate from downstream to upstream, and results in demand planning errors and this is followed by inaccurate forecasting. This situation is referred to as the bullwhip effect phenomenon. Demand management is also determined based on either a centralized or decentralized distribution strategy. Both strategies will influence the accuracy of demand planning and its effect on the bullwhip phenomenon. Unfortunately, research investigating the relationship between centralization and decentralization strategies for demand and the bullwhip effect is still limited. To answer this shortcoming, this paper has two scenarios for its analysis. First, forecasting is done to determine the accuracy of demand planning, indicated by the smallest forecast error value. Based on the results of the analysis, it is known that single exponential smoothing with alpha 0.5 is the best forecasting method. The second step is to measure the bullwhip effect; the results show that the coefficient is less than one. This coefficient indicated that the company underperformed in the fulfillment of its customers’ needs. Theoretically, this paper extends the literature on demand management in the supply chain by considering centralized and decentralized strategies.

A Combined Model Based on EOBL-CSSA-LSSVM for Power Load Forecasting

Inaccurate electricity load forecasting can lead to the power sector gaining asymmetric information in the supply and demand relationship. This asymmetric information can lead to incorrect production or generation plans for the power sector. In order to improve the accuracy of load forecasting, a combined power load forecasting model based on machine learning algorithms, swarm intelligence optimization algorithms, and data pre-processing is proposed. Firstly, the original signal is pre-processed by the VMD–singular spectrum analysis data pre-processing method. Secondly, the noise-reduced signals are predicted using the Elman prediction model optimized by the sparrow search algorithm, the ELM prediction model optimized by the chaotic adaptive whale algorithm (CAWOA-ELM), and the LSSVM prediction model optimized by the chaotic sparrow search algorithm based on elite opposition-based learning (EOBL-CSSA-LSSVM) for electricity load data, respectively. Finally, the weighting coefficients of the three prediction models are calculated using the simulated annealing algorithm and weighted to obtain the prediction results. Comparative simulation experiments show that the VMD–singular spectrum analysis method and two improved intelligent optimization algorithms proposed in this paper can effectively improve the prediction accuracy. Additionally, the combined forecasting model proposed in this paper has extremely high forecasting accuracy, which can help the power sector to develop a reasonable production plan and power generation plans.

A study of the reliability of cross-sectional earnings forecasting models for estimating IPO firms’ implied cost of capital

PurposeThis study aims to evaluate the earnings forecasting models of Hou et al. (J Account Econ, 53:504–526, 2012) and Li and Mohanram (Rev Account Stud, 19:1152–1185, 2014) in terms of bias and accuracy and validity of the implied cost of capital (ICC) estimates for a sample of initial public offerings (IPOs).Design/methodology/approachThe authors use a sample of 1,657 NYSE, Amex and Nasdaq IPOs from 1972 to 2013.FindingsThe models of Hou et al. and Li and Mohanram produce relatively inaccurate and biased earnings forecasts, leading to unreliable ICC estimates, particularly for small and loss-making IPOs that constitute the bulk of new listings. As a remedy, the authors propose a new earnings forecasting model, a combination of Hou et al.’s and Li and Mohanram’s earnings persistence models, and show that it produces more accurate and less biased earnings forecasts and more valid ICC estimates.Originality/valueThe study contributes novel results to the literature on the validity of cross-sectional earnings models in forecasting IPO firm earnings and estimating the ICC. The findings are directly relevant for practitioners, who can improve their earnings forecasting accuracy for IPO firms and related ICC estimates. The insights can be extended to other settings where investors have limited access to financial information, such as acquisitions of private targets.

Federated learning for IoE environments: A service provider revenue maximization framework

In accordance with the Internet of Everything (IoE) paradigm, millions of people and billions of devices are expected to be connected to each other, giving rise to an ever increasing demand for application services with a strict quality of service requirements. Therefore, service providers are dealing with the functional integration of the classical cloud computing architecture with edge computing networks. However, the intrinsic limited capacity of the edge computing nodes implies the need for proper virtual functions' allocations to improve user satisfaction and service fulfillment. In this sense, demand prediction is crucial in services management and exploitation. The main challenge here consists of the high variability of application requests that result in inaccurate forecasts. Federated learning has recently emerged as a solution to train mathematical learning models on the users' site. This paper investigates the application of federated learning to virtual functions demand prediction in IoE based edge cloud computing systems, to preserve the data security and maximise service provider revenue. Additionally, the paper proposes a virtual function placement based on the services demand prediction provided by the federated learning module. A matching based tasks allocation is proposed. Finally, numerical results validate the proposed approach, compared with a chaos theory prediction scheme.

On the Application of ARIMA and LSTM to Predict Order Demand Based on Short Lead Time and On-Time Delivery Requirements

Suppliers are adjusting from the order-to-order manufacturing production mode toward demand forecasting. In the meantime, customers have increased demand uncertainty due to their own considerations, such as end-product demand frustration, which leads to suppliers’ inaccurate demand forecasting and inventory wastes. Our research applies ARIMA and LSTM techniques to establish rolling forecast models, which greatly improve accuracy and efficiency of demand and inventory forecasting. The forecast models, developed through historical data, are evaluated and verified by the root mean squares and average absolute error percentages in the actual case application, i.e., the orders of IC trays for semiconductor production plants. The proposed ARIMA and LSTM are superior to the manufacturer’s empirical model prediction results, with LSTM exhibiting enhanced performance in terms of short-term forecasting. The inventory continued to decline significantly after two months of model implementation and application.

Chance constrained unit commitment approximation under stochastic wind energy

With the increasing integration of renewable energy into power networks, the scheduling and operation of power systems become more challenging, in particular because of the intermittency of wind speed. Thus, appropriate optimization tools are needed to efficiently manage the operation of modern power systems. This paper considers the unit commitment problem of power systems under high wind energy penetration. To mitigate the potential negative effects of inaccurate wind energy forecast, a chance constrained model is employed to guarantee that supply and demand for energy be balanced at each time period over a finite planning horizon. As joint chance constraint is known to be difficult to handle analytically, two approximation approaches, namely quantile- and p-efficiency-based, are proposed and benchmarked against the well-known scenario-based approximation. The results indicate that the proposed schemes provide feasible solutions and tight lower bounds of the true operation cost within a much shorter time frame than the scenario-based counterpart.

Leveraging Artificial Intelligence for Sustainability in the Textile and Fashion Industry

Environmental problems have posed serious threats to society. Given the enormous environmental footprints created by the fashion industry, fashion companies have developed strategic plans to achieve sustainability goals. While using biodegradable or recycled materials helps the environment, this article contends that the major environmental problems are caused by inaccurate demand forecasting and excess production. It discusses the role of artificial intelligence (AI) in improving demand prediction and supply chain performance using Bestseller India and IBM’s AI products as examples.

Coherent Mortality Model in A State-Space Approach

Mortality improvements that have recently become apparent in most developing countries have significantly shaped queries on forecast divergent between populations in recent years. Therefore, to ensure a more coherent way of forecasting, previous researchers have proposed multi-population mortality model in the form of independent estimation procedures. However, similar to single-population mortality model, such independent approaches might lead to inaccurate prediction interval. As a result of this inaccurate mortality forecasts, the life expectancies and the life annuities that the mortality model aims to generate is underestimated. In this study, we propose another new extension of the multi-population mortality model in a joint estimation approach by recasting the model into a state-space framework. A combination of augmented Li-Lee and O’Hare-Li methods are employed, before we transform the proposed model into a state-space formulation. In addition, this study incorporates the quadratic age effect parameter to the proposed model to better capture the younger ages mortality. We apply the method to gender and age-specific data for Malaysia. The results show that our latter framework brings a significant contribution to the multi-population mortality model due to the incorporation of joint-estimate and quadratic age effect parameters into the model’s structure. Consequently, the proposed model improves the mortality forecast accuracy.

Regional GDP Distortion and Analyst Forecast Accuracy: Evidence from China

The reliability of China’s GDP data has been questioned for a long time. Prior studies have discussed the causes of GDP distortion in China, but the evidence on the economic consequences is scant. This paper examines the economic consequences of regional GDP distortion from the perspective of analyst forecasts. We find that regional GDP distortion leads to lower analyst forecast accuracy, and this result is robust to potential endogeneity. Further investigations show that analysts with information advantages have the ability to resist the distorted GDP data and issue accurate forecasts. Political pressure is another factor leading to inaccurate analyst forecasts. Our paper contributes to the literature by highlighting the importance of the reliability of GDP figures as a determinant of analyst forecast accuracy.

EV Scheduling Framework for Peak Demand Management in LV Residential Networks

Increased electrification in the residential and transport sectors is changing the energy demand profiles significantly, which results in reshaped peak demand. These changes in demand profiles can cause grid overloading and jeopardize network reliability especially when the excessive use of electricity within a network is uncoordinated. In this article, an aggregated coordination mechanism is proposed for electric vehicle (EV) charge–discharge scheduling to manage the peak demand in the low-voltage (LV) residential networks. The proposed model uses mixed-integer-programming-based optimization approach to minimize the cost of energy while managing the peak demand and complying with grid constraints. A stochastic model is presented to account for the uncertainties associated with forecast inaccuracies of the day-ahead scheduling. The proposed strategy is assessed by means of simulation studies considering an LV residential neighborhood in Sydney, Australia. The results indicate the effectiveness of the proposed strategy to minimize the cost of electricity for the EV owners while managing the peak demand for the grid operators. Comparison with the state-of-the-art EV scheduling strategies indicates that the proposed strategy can improve the load factor of the local network up to 36%, the peak-to-average ratio up to 27%, and cost reductions up to 56%.

New methodologies for predicting corridor travel time mean and reliability

ABSTRACT Accurate travel time prediction is very important for real-time traveller information systems. Many existing traveller information systems provide point estimates of forecast travel times. Often the forecast corridor travel time is estimated as a direct summation of the forecast link travel times on the route. This approach neglects the correlation between link travel times and may lead to inaccurate route travel time forecasts. This paper improves upon the simple addition method by accounting for the dependency of link travel times on the arrival time at that specific link which further correlates to its preceding links. In addition, this paper also explores the potential of using the nonlinear autoregressive with exogenous inputs (NARX) model and feedforward neural network model to forecast the corridor travel time mean and reliability metrics. To the authors knowledge this is the first time, short-term travel time reliability is measured by a reliability interval which is based on the forecasts of corridor travel time mean and standard deviation. The prediction methodologies developed in this paper are tested on an urban arterial that has been instrumented with Bluetooth readers so empirical travel times are available. It was found that the proposed NARX model outperforms the other models that were studied with respect to mean corridor travel time prediction. In terms of the reliability interval prediction, the performance of various models is presented as a Pareto Optimal Frontier trading off accuracy and usability. The proposed NARX model and three other tested models are all on the Pareto Optimal Frontier. Highlights Explores the nonlinear autoregressive with exogenous inputs (NARX) model for travel time predictions; Develops models to forecast urban arterial corridor travel time means and reliability metrics; Develops forecasting models with link-based or corridor-based travel time inputs, respectively; Evaluates the performance of various models using a Pareto Optimal Frontier; Compared travel time reliability metrics: reliability interval, 95th percentile, and buffer index.