Optimal Forecast Research Articles

PERAMALAN INDEKS HARGA SAHAM GABUNGAN (IHSG) MENGGUNAKAN METODE BACKPROPAGATION DAN RECURRENT NEURAL NETWORK (RNN)

Forecasting is an objective calculation and uses past data to predict something in the future. The economic activity of buying and selling shares is a common activity in this digital era. Investment in the capital market is currently popular among the general public. Forecasting in the digitalization era has undergone many developments, as well as its application not only in agriculture, but also in economics, politics and other fields. The development of forecasting in the economic field is growing rapidly with the use of more complex mathematical models and the application of information technology. Forecasting is an objective calculation and uses past data to predict something in the future. Utilization of artificial neural network (ANN) is one of the solutions for forecasting with more accurate results. One of the algorithms in ANN is the backpropagation algorithm. Apart from using other backpropagation forecasting methods that can be used is the deep learning (DL) method with a recurrent neural network algorithm. The backpropagation method produces an optimal MAPE of 13.28%. with an architectural model of seven input layer neurons, seven hidden layer neurons, and using the sigmoid activation function. The recurrent neural network method produces an optimal MAPE of 1.12%. with an architectural model of seven input layer neurons, ten hidden layer neurons, and using the tanh activation function. If the two optimal MAPEs of each method are compared, the Recurrent Neural Network Method produces more optimal forecasting compared to the backpropagation method.

A novel integrated method for improving the forecasting accuracy of crude oil: ESMD-CFastICA-BiLSTM-Attention

The volatility of crude oil price can significantly impact the stability of the crude oil market and even the global economy. Effectively predicting global crude oil price and volatility provides a scientific basis for decision-making for market regulators and investors worldwide, promoting the sound development of the global economy. In this study, we integrate signal processing with deep learning methods to present an optimal forecasting strategy for global crude oil price and volatility. We select the daily price of the WTI crude oil market from April 4, 1983, to December 12, 2023, for calculating volatility. Subsequently, employing extreme-point symmetric empirical mode decomposition (ESMD), K-means clustering, and fast independent component analysis method, we decompose and reconstruct the forecasting data, obtaining independent components with non-Gaussian characteristics. These components serve as inputs to estimate the accuracy of various models, including BiLSTM, Attention, LSTM, SVR, RF, and their combinations, in predicting crude oil price and volatility from both a point prediction and interval prediction perspective. Empirical results demonstrate that data decomposition, reconstruction, and the BiLSTM-Attention model outperform other models in predicting crude oil price and volatility.

Are USDA Forecasts Optimal? A Systematic Review

AbstractThe goal of this paper is to systematically review the literature on United States Department of Agriculture (USDA) forecast evaluation and critically assess their methods and findings. The fundamental characteristics of optimal forecasts are bias, accuracy and efficiency as well as encompassing and informativeness. This review revealed that the findings of these studies can be very different based on the forecasts examined, commodity, sample period, and methodology. Some forecasts performed very well, while others were not very reliable, resulting in forecast specific optimality record. We discuss methodological and empirical contributions of these studies as well as their shortcomings and potential opportunities for future work.

Exponential smoothing method against the gradient boosting machine learning algorithm-based model for materials forecasting to minimize inventory

The optimization of resources and reduction of costs through efficient inventory management are paramount to organizational success. This study undertakes a comparative analysis of two distinct forecasting methodologies, Exponential Smoothing (ES) and Gradient Boosting (GB), within the framework of materials forecasting aimed at inventory minimization. Our study introduces innovation by methodically scrutinizing these approaches within a unified framework, shedding light on their merits and shortcomings. This comparative analysis gives practitioners a practical roadmap for the optimal forecasting strategy to streamline inventory management operations. Methodologies are evaluated based on their efficiency in predicting material demand, encompassing metrics such as accuracy, computational efficiency, and suitability across various inventory management scenarios. Response surface methodology entails refining processes to modify factorial variables’ configurations to attain a desired peak or trough in response. The SPSS results show that the ES method has 43.20%, surpassing the accuracy of the inventory optimization model, which stood at 65.08%. The response surface methodology results show that 45.20% profit was achieved for the variable and operational cost process parameters. This research seeks to unveil the traces of each method, facilitating decision-makers in selecting an optimal forecasting strategy tailored to their specific inventory management requirements. The analysis shows that the ES method surpasses the accuracy of the GB machine learning for material forecasting to minimize inventory.

Study on visibility forecast optimization based on aerosol- meteorological feedbacks in wet conditions

Study on visibility forecast optimization based on aerosol- meteorological feedbacks in wet conditions

An Accident Prediction Model Based on ARIMA in Kuala Lumpur, Malaysia, Using Time Series of Actual Accidents and Related Data

Recently, there has been an emerging trend to analyse time series data and utilise sophisticated tools for optimally fitting time series models. To date, Malaysian industrial accident data is underutilised and lacks informative records. Thus, this paper aims to investigate the Malaysian accident database and further evaluate the optimal forecasting models in accident prediction. The model’s input was based on available data from the Department of Occupational Safety and Health, Malaysia (DOSH), from 2018 until 2021, with 80% of the dataset to train the models and the remaining 20% for validation. The negative binomial and Poisson distribution prediction showed a mean absolute percentage error (MAPE) of 33% and 51%, respectively. It indicated that the negative binomial performed better than the Poisson distribution in accident frequency prediction. The available time series accident data were gathered for four years, and stationarity was checked in R Studio software for the Augmented Dickey-Fuller test. The lowest Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC) and other error values were used to justify the best model, which was the ARIMA(2,0,2)(2,0,0)(12) model. The ARIMA models were considered after the data showed autocorrelation. The MAPE for both ARIMA in R and manual time series were 40% and 49%, respectively. Therefore, the accident prediction by using R Studio would outperform the manually negative binomial and Poisson distribution. Based on the findings, industrial safety practitioners should report accidents to DOSH truthfully in the era of digitalisation. It could enable future data-driven accident predictions to be carried out.

Conditionally optimal weights and forward-looking approaches to combining forecasts

In forecasting, there is a tradeoff between in-sample fit and out-of-sample forecast accuracy. Parsimonious model specifications typically outperform richer model specifications. Consequently, information is often withheld from a forecast to prevent over-fitting the data. We show that one way to exploit this information is through forecast combination. Optimal combination weights in this environment minimize the conditional mean squared error that balances the conditional bias and the conditional variance of the combination. The bias-adjusted conditionally optimal forecast weights are time varying and forward looking. Real-time tests of conditionally optimal combinations of model-based forecasts and surveys of professional forecasters show significant gains in forecast accuracy relative to standard benchmarks for inflation and other macroeconomic variables.

A hybrid deep learning model with an optimal strategy based on improved VMD and transformer for short-term photovoltaic power forecasting

The precise forecasting of photovoltaic (PV) power is important for efficient grid management. To enhance the analysis and processing capability of PV characteristics, address the feature extraction challenges for long sequences, and improve forecasting accuracy, this study presents a robust hybrid deep learning model for PV power forecasting. First, a dynamic mean pre-processing algorithm is applied for data cleaning. Subsequently, an improved whale variational mode decomposition (IWVMD) algorithm is proposed for data decomposition in multichannel multi-scale modeling. Furthermore, a novel context-embedded causal convolutional Transformer (CCTrans) structure is used to predict each subsequence, and an optimal strategy is formulated for both input and output under the combined dynamic contextual information and single target variable forecasting (CDCTF) pattern. Finally, the forecasting results are reconstructed. Experiments are conducted to evaluate the performance of the model across different seasons, using publicly available datasets from the Desert Knowledge Australia Solar Center (DKASC). Ablation studies, validation with diverse datasets, and comparisons with other models confirm the effectiveness, accuracy, robustness, and generalizability of the model. In addition, recommendations for optimal forecasting ranges for different seasons are provided.

A forecasting model for oil prices using a large set of economic indicators

AbstractThis paper examines the predictability of the changes in Brent oil futures prices using a multilayer perceptron artificial neural network that exploits the information contained in the largest possible set of economic indicators. Feature engineering is employed to identify the most important predictors of the change in Brent oil futures prices. We find that oil‐market‐specific variables are important predictors. Our findings also suggest that forecasts of the change in the Brent oil futures prices from the multilayer perceptron that exploits the informational content of all and oil‐market‐specific predictors exhibit higher statistical forecast accuracy than the random walk. Tests of forecast optimality indicate that the forecasts generated using oil‐market‐specific predictors are optimal. We discuss the policymaking and practical relevance of our results.

On the optimal forecast with the fractional Brownian motion

This paper investigates the performance of different forecasting formulas with fractional Brownian motion based on discrete and finite samples. Existing literature presents two formulas for generating optimal forecasts when continuous records are available. One formula relies on a history over an infinite past, while the other is designed for a record limited to a finite past. In reality, only observations at discrete time points over a finite past are available. In this case, the forecasting formula, which has been widely used in the literature, is the one obtained by Gatheral et al. (Volatility is rough. Quant. Finance, 2018, 18(6), 933–949) that truncates and discretizes the formula based on continuous records over an infinite past. The present paper advocates an alternative forecasting formula, which is the conditional expectation based on finite past discrete-time observations. The findings suggest that the conditional expectation approach produces more accurate forecasts than the existing method, as demonstrated by both simulated data and actual daily realized volatility (RV) observations. Moreover, we also provide empirical evidence showing that the conditional expectation approach can lead to larger economic values than the existing method.

Empowering urban climate resilience and adaptation: Crowdsourcing weather citizen stations-enhanced temperature prediction

The growing impact of climate change, including extreme weather events, represents a significant challenge for humanity. With most of the world's population living in urban areas, the urban heat island effect and anthropogenic heat contribute to elevated city temperatures. This increase in urban warming threatens human health and demands a deeper understanding of thermal distribution in urban environments. Collecting accessible and widespread temperature data in urban areas is essential to address this challenge. This study aims to develop a methodology for anticipating temperature distribution in urban environments, leveraging Citizen Weather Stations (CWS) as valuable crowdsourcing data sources. The ultimate goal is to create a predictive model that estimates urban temperatures based on government meteorological station forecasts, improving urban planning, regulating temperature-based routes, preventing health issues in vulnerable populations, and enhancing urban livability. The methodology is divided into three fundamental stages: data acquisition through CWS with citizen collaboration, the development and evaluation of optimal forecast models based on government weather stations (SWS) data, and its exploitation in terms of utility and applicability. This methodology encompasses data collection and filtering to ensure its usefulness and implement reliable models. The resulting tool facilitates informed decision-making and precise seasonal event planning in urban environments, effectively addressing the challenges of climate extrapolation and contributing to more effective adaptation and mitigation strategies in climate change and heatwaves. The results obtained probe the feasibility of using CWS to predict temperatures in urban environments, which has been demonstrated accurately. This is a significant achievement, as CWS has proven to be a reliable source of climate data for this context. Also, the filtering process described and applied to the case study has proven effective, discarding approximately 34.87 % of the data. This is achieved by detecting and eliminating anomalies, considering station availability, and adhering to specific quality criteria. Finally, the developed prediction model has demonstrated its ability to optimally estimate urban temperatures, utilizing climate prediction data provided by government weather stations (SWS). The model performance indicators support this claim. For the linear regression model, a Mean Squared Error (MSE) of 2.177 and an R-squared (R2) of 0.960 are obtained, while for the neural network, an MSE of 1.284 and an R2 of 0.976 are achieved.

The influence of immediate intraoperative loading with a splinting component on supporting tissues during a one-stage implant.

Aim: To study the specifics of the impact of immediate intraoperative loading with a splinting component on supporting tissues during a one-stage implantation protocol. Materials and Methods: In the course of the study, orthopedic treatment was carried out for 55 patients aged 29 to 60 years. The following were performed: cone-beam computed tomography, software planning and intraoral scanning with an optical scanner, one-stage implantation protocol, assessment of implant stability with the Osstell ISQ device, microcirculation study in the peri-implant area using laser Doppler flowmetry (LDF). Results: It was established that around loaded implants there is an increase in blood flow and vasomotor activity of the microcirculatory channel of the supporting tissues, an increase in the volume of bone tissue and an increase in torque, which is the optimal forecast for the acceleration of the pace of osseointegration. Conclusions: The use of a splinting component during immediate intraoperative functional masticatory load accelerates the dynamics of bone tissue remodeling processes around the implant, which is an optimal prediction of osseointegration rates in various dental implantation protocols and is consistent with high values of the implant stability coefficient.

A hybrid model integrating artificial neural network with multiple GARCH-type models and EWMA for performing the optimal volatility forecasting of market risk factors

The 2008 financial crisis has highlighted the lack of precision in the market risk metrics that financial institutions must report to the regulator. The use of Machine Learning techniques in stock markets and the treasury (Front–Back Office) of financial institutions is a key tool for optimizing its own resources, internal processes and risk measures. We propose a hybrid methodology to better capture the volatility of market risk factors with Value-at-risk models in periods of stress, but also in periods of stability compared to traditional metrics. This hybrid model uses different types of artificial neural networks and traditional metrics to perform the optimal forecast of volatility applied to the main market risk nodes of the Spanish stock market. We use data from the following main market risk factors: returns on Santander Bank shares, the Spanish Stock Market Index, Euro/Dollar exchange rates and the index that measures the total return performance of a funded long credit position in the on-the-run iTraxx Crossover 5-Year-Index. Our contribution is a hybrid model that combines correct sequential pattern learnings with an improved prediction performance in the volatility of market risk factors. Our findings show that the Support Vector Machine and the Long-Short-Term Memory Model present better prediction results in all factors in the stability periods. Therefore, the proposed method is promising for application in risk management systems.

The impact of Indian radar and lightning data assimilation on the short‐range forecasts of heavy rainfall events

AbstractThe accurate prediction of high‐impact weather systems using cloud‐resolving models is still a challenge among researchers. This study evaluates the consistency of the combination of the three‐dimensional variational technique within the Gridpoint Statistical Interpolation assimilation system (GSI‐3DVAR) and nudging in the same modelling system on short‐range forecasts of three heavy rainfall events from the southwest monsoon season of 2021 over the Indian subcontinent. Three experiments have been conducted (i) control (CNTL): assimilated conventional and satellite observations; (ii) radar and lightning data assimilation (RLDA): assimilated radar reflectivity and lightning proxy reflectivity data along with all observations used in CNTL; and (iii) lightning data assimilation (LDA): same as RLDA but without the assimilation of radar data; particularly done to test the impact of assimilation of only lightning data. The model‐simulated rainfall is evaluated by using the Integrated Multi‐Satellite Retrievals (IMERG) for Global Precipitation Measurement (GPM IMERG) rainfall data. The intercomparison of LDA and RLDA for event 1 highlighted that both represent the convective regions reasonably better than CNTL, but RLDA outperforms LDA and thus further assimilation experiments are done with RLDA. RLDA provided reasonably accurate forecasts compared to CNTL, which is evident in the spatial distribution of rainfall and area‐averaged three‐hourly accumulated rainfall. Verification metrics for the three selected heavy rainfall events reveal that an optimal forecast performance (especially in the first six hours of free forecast) is obtained by the simulation with assimilation of radar and lightning data during the pre‐forecast period, through correcting the position and timing of convective centres. The probability of detection (POD) values are higher for light rainfall categories than for the heavy rain categories. POD values were higher in RLDA than CNTL throughout simulation for all three events. For all these three selected events, fractions skill scores (FSS) of RLDA are always better than CNTL with different neighbourhood sizes for different threshold values throughout the forecast period.

Reservoir operations under uncertainty with moving-horizon approach and ensemble forecast optimization

Inflow forecasts are a basic pre-requisite to support decisions for anticipating reservoir operations. The fact that hydrological forecasts are uncertain, noting that reservoirs often fulfill multiple purposes, bears the risk of non-ideal reservoir operation. Optimization techniques can help to identify the best operational scheme under given inflow. To address the forecast uncertainty, operators can repeat optimization (moving horizon approach) and optimize for inflow forecast ensembles. This article aims to contribute to a better understanding of how the different methods work, how to interpret the results and what the effect of user choices on the optimization results is. The moving-horizon approach and three ensemble optimization methods were applied on a hydropower reservoir in Norway under flood conditions. The functional principle of each method is explained, and advantages and drawbacks of the different methods are discussed with the help of performance indicators.

A new rolling forecasting framework using Microsoft Power BI for data visualization: A case study in a pharmaceutical industry

Demand forecasting is a vital step for production planning and consequently, for supply chain efficiency, especially for the pharmaceutical (pharma) supply chain due to its unique characteristics. Numerous models and techniques that are proposed in the literature but little in concrete and generic framework to forecasting process, mainly for pharmaceutical supply chain. Unlike studies in the literature, this study not only perfectly predict the sales of a pharma manufacturer, but also visualize the results via a developed dashboard using modern information technology and business intelligence. In this research, a rolling forecasting framework comprising of different steps and specialized tools is proposed that can assist supply chain managers to perform an accurate sales forecasting and consequently a better performance and specifically patient satisfaction. The proposed generic framework combines the use of Visual studio C++ software to extract optimal forecasting and the Power BI software to monitor the accuracy of the obtained sales forecasts. Three exponential smoothing methods are integrated in the proposed framework, which is open to adding more new forecasting methods. The proposed framework is tested for many data sets from a pharmaceutical manufacturer company, and the results obtained show superior performance, especially a clear decline in both forecast errors, which can reach 75% and a drop of stock level to 50%. Therefore, the company is currently using it and a future integration with their ERP is being carried out. The proposed rolling forecasting framework contributes to insightful decision-making through the visualization of accurate future sales and turnover, and consequently, an efficient stock management and effective production planning.

An adaptive multi-objective optimal forecast combination and its application for predicting intermittent demand

While time series forecasting models are generally trained by optimising certain forms of error, the end-user’s forecasting needs in a multi-objective setting can be broader, and often mutually conflicting. A production manager may prioritise high product fill rates and low average inventory resulting from a forecast over just low error. The conflict among multiple objectives is notably worrisome in intermittent demand forecasting, where error-minimising approaches can devalue the practitioner’s objectives. To address such forecasting problems, we propose an Adaptive Multi-objective Optimal Combination (AMOC) of forecasts which incorporates the end-user’s preferences across multiple objectives. We demonstrate the use of AMOC in a real-life application of intermittent demand forecasting for optimising four distinct inventory management objectives using five specialised forecasting methods across single-period and multi-period inventory handling scenarios. Additionally, we conduct a comprehensive experiment on a subset of M5 competition data to exhibit the robustness of the AMOC using 13 diverse forecasting methods and four statistical objectives.

Spatio-Temporal Alignment and Track-To-Velocity Module for Tropical Cyclone Forecast

The prediction of a tropical cyclone’s trajectory is crucial for ensuring marine safety and promoting economic growth. Previous approaches to this task have been broadly categorized as either numerical or statistical methods, with the former being computationally expensive. Among the latter, multilayer perceptron (MLP)-based methods have been found to be simple but lacking in time series capabilities, while recurrent neural network (RNN)-based methods excel at processing time series data but do not integrate external information. Recent works have attempted to enhance prediction performance by simultaneously utilizing both time series and meteorological field data through feature fusion. However, these approaches have relatively simplistic methods for data fusion and do not fully explore the correlations between different modalities. To address these limitations, we propose a systematic solution called TC-TrajGRU for predicting tropical cyclone tracks. Our approach improves upon existing methods in two main ways. Firstly, we introduce a Spatial Alignment Feature Fusion (SAFF) module to address feature misalignment issues in different dimensions. Secondly, our Track-to-Velocity (T2V) module leverages time series differences to integrate external information. Our experiments demonstrate that our approach yields highly accurate predictions comparable to the official optimal forecast for a 12 h period.

Forecasting crude oil futures price using machine learning methods: Evidence from China

Crude oil is an indispensable energy resource. With the establishment of the local crude oil futures market in China, providing accurate forecasts for crude oil futures price is urgent. To cope with this challenge, this paper introduces artificial intelligence methods to evaluate the optimal forecasting strategy for China's crude oil futures price. We use machine learning to process historical information, volatility and non-linear features. Using daily data from March 26, 2018 to February 28, 2023, we estimate the forecasting effects of RNN, LSTM, GRU, SVR, MLP, CNN and BP models on China crude oil futures, respectively. With a series of evaluation tests, we demonstrate that the GRU model outperforms other models in terms of forecast accuracy and performance for China's crude oil futures price. Taking multiple influencing factors into account, the forecasting accuracy of proposed models is improved by including influential factors. Therefore, these findings effectively explore the forecasting of China's crude oil futures prices, contributing to the improvement of the emerging crude oil futures market and the management of energy price risks.

Volatility predictability in crude oil futures: Evidence based on OVX, GARCH and stochastic volatility models

The paper examines the volatility predictive ability of the CBOE crude oil volatility index (OVX), GARCH and Stochastic Volatility Models in the crude oil market. Specifically, the dynamics of two major crude oil pricing benchmarks — Brent in Europe and WTI in America are compared. OVX index is able to provide the optimal forecast for the volatility of Brent’s future. The GJR-GARCH model outperforms its competition volatility models in both oil markets. It also reveals that WTI exhibits a faster and stronger response to market shocks compared to Brent oil. Most importantly, the predictability exhibit consistent and convincing results during remarkable events: the 2014 oil price decline, the 2020 coronavirus pandemic, and the 2022 Russian–Ukraine war. In practice, the paper contributes to portfolio strategy construction and investment risk management.