Recurrent ensemble random vector functional link neural network for financial time series forecasting

Applied Soft Computing, Год журнала: 2024, Номер 161, С. 111759 - 111759

Опубликована: Май 16, 2024

Financial time series forecasting is crucial in empowering investors to make well-informed decisions, manage risks effectively, and strategically plan their investment activities. However, the non-stationary non-linear characteristics inherent data pose significant challenges when accurately predicting future forecasts. This paper proposes a novel Recurrent ensemble deep Random Vector Functional Link (RedRVFL) network for financial forecasting. The proposed model leverages randomly initialized fixed weights recurrent hidden layers, ensuring stability during training. Furthermore, incorporating stacked layers enables representation learning, facilitating extraction of complex patterns from data. generates forecast by combining outputs each layer through an approach. A comparative analysis was conducted against several state-of-the-art models over time-series datasets, results demonstrated superior performance our terms accuracy predictive capability.

Язык: Английский

---

An innovative interpretable combined learning model for wind speed forecasting

Applied Energy, Год журнала: 2024, Номер 358, С. 122553 - 122553

Опубликована: Янв. 5, 2024

Язык: Английский

---

A multi-scale analysis method with multi-feature selection for house prices forecasting

Applied Soft Computing, Год журнала: 2025, Номер unknown, С. 112779 - 112779

Опубликована: Янв. 1, 2025

Язык: Английский

---

Biochar price forecasting: A novel methodology for enhancing market stability and economic viability

Journal of Environmental Management, Год журнала: 2025, Номер 377, С. 124681 - 124681

Опубликована: Фев. 26, 2025

Язык: Английский

---

Operational Research: methods and applications

Journal of the Operational Research Society, Год журнала: 2023, Номер 75(3), С. 423 - 617

Опубликована: Дек. 27, 2023

Throughout its history, Operational Research has evolved to include methods, models and algorithms that have been applied a wide range of contexts. This encyclopedic article consists two main sections: methods applications. The first summarises the up-to-date knowledge provides an overview state-of-the-art key developments in various subdomains field. second offers wide-ranging list areas where applied. is meant be read nonlinear fashion used as point reference by diverse pool readers: academics, researchers, students, practitioners. entries within applications sections are presented alphabetical order. authors dedicate this paper 2023 Turkey/Syria earthquake victims. We sincerely hope advances OR will play role towards minimising pain suffering caused future catastrophes.

Язык: Английский

---

A Multi‐Model Ensemble of Baseline and Process‐Based Models Improves the Predictive Skill of Near‐Term Lake Forecasts

Water Resources Research, Год журнала: 2024, Номер 60(3)

Опубликована: Март 1, 2024

Abstract Water temperature forecasting in lakes and reservoirs is a valuable tool to manage crucial freshwater resources changing more variable climate, but previous efforts have yet identify an optimal modeling approach. Here, we demonstrate the first multi‐model ensemble (MME) reservoir water forecast, method that combines individual model strengths single framework. We developed two MMEs: three‐model process‐based MME five‐model includes empirical models forecast profiles at temperate drinking reservoir. found improved performance by 8%–30% relative MME, as quantified using aggregated probabilistic skill score. This increase was due large improvements bias despite increases uncertainty. High correlation among resulted little improvement models. The utility of MMEs highlighted results: (a) no performed best every depth horizon (days future), (b) avoided poor performances rarely producing worst for any forecasted period (<6% ranked forecasts over time). work presents example how existing can be combined improve discusses value utilizing MMEs, rather than models, operational forecasts.

Язык: Английский

---

Forecast combination using grey relational analysis and Choquet fuzzy integral for container throughput forecasting

Expert Systems with Applications, Год журнала: 2024, Номер 252, С. 124170 - 124170

Опубликована: Май 6, 2024

Язык: Английский

---

COVID-19 and Rates of Cancer Diagnosis in the US

JAMA Network Open, Год журнала: 2024, Номер 7(9), С. e2432288 - e2432288

Опубликована: Сен. 6, 2024

Importance US cancer diagnoses were substantially lower than expected during the COVID-19 pandemic in 2020. A national study on extent to which rates recovered 2021 has not yet been conducted. Objective To examine observed vs rate trends for January 2020 December 2021. Design, Setting, and Participants This cross-sectional, population-based of incidence used Surveillance, Epidemiology, End Results 22 (SEER-22) Registries Database, covers 47.9% population. Included individuals those with an invasive diagnosis reported registries included SEER-22 between 1, 2000, 31, Exposures Age, sex, race ethnicity, urbanicity, stage at diagnosis. Main Outcomes Measures Expected measured years from prepandemic using ensemble forecasting methods. Relative difference numbers potentially missed cases measured. The 1 578 697 2021, including 798 765 among male (50.6%) 909 654 persons aged 65 or older (57.6%). Observed all-sites by 9.4% (95% prediction interval [PI], 8.5%-10.5%), 2.7% PI, 1.4%-3.9%), 6.0% across both combined 5.1%-7.1%), resulting 149 577 undiagnosed 126 059-176 970). Of 4 screening-detected cancers, only female breast showed significant recovery exceeding 2.5% 0.1%-4.8%), while reductions remained lung (9.1% expected; 95% 6.4%-13.2%) cervical (4.5% 0.4%-8.0%), particularly early Rates returned individuals, younger years, non-Hispanic Asian Pacific Islander ethnicity. Conclusions Relevance cross-sectional found that improved but continued be expected, adding existing deficit diagnosed Particular attention should directed strategies immediately increase screenings make up lost ground.

Язык: Английский

---

A review of machine learning concepts and methods for addressing challenges in probabilistic hydrological post-processing and forecasting

Frontiers in Water, Год журнала: 2022, Номер 4

Опубликована: Окт. 5, 2022

Probabilistic forecasting is receiving growing attention nowadays in a variety of applied fields, including hydrology. Several machine learning concepts and methods are notably relevant towards addressing the major challenges formalizing optimizing probabilistic implementations, as well equally important challenge identifying most useful ones among these implementations. Nonetheless, practically-oriented reviews focusing on such methods, how can be effectively exploited above-outlined essential endeavour, currently missing from hydrological literature. This absence holds despite pronounced intensification research efforts for benefitting this same It also substantial progress that has recently emerged, especially field post-processing, which traditionally provides hydrologists with Herein, we aim to fill specific gap. In our review, emphasize key ideas information lead effective popularizations, an emphasis support successful future implementations further scientific developments. forward-looking direction, identify open questions propose explored future.

Язык: Английский

---

A user-centered explainable artificial intelligence approach for financial fraud detection

Finance research letters, Год журнала: 2023, Номер 58, С. 104309 - 104309

Опубликована: Авг. 9, 2023

Язык: Английский

---