A Hierarchical RF-XGBoost Model for Short-Cycle Agricultural Product Sales Forecasting DOI Creative Commons
Jiawen Li,

Binfan Lin,

Wang Peixian

и другие.

Foods, Год журнала: 2024, Номер 13(18), С. 2936 - 2936

Опубликована: Сен. 17, 2024

Short-cycle agricultural product sales forecasting significantly reduces food waste by accurately predicting demand, ensuring producers match supply with consumer needs. However, the is often subject to uncertain factors, resulting in highly volatile and discontinuous data. To address this, a hierarchical prediction model that combines RF-XGBoost proposed this work. It adopts Random Forest (RF) first layer extract residuals achieve initial results based on correlation features from Grey Relation Analysis (GRA). Then, new feature set residual clustering generated after applied classify characteristics of residuals. Subsequently, Extreme Gradient Boosting (XGBoost) acts as second utilizes those yield results. The final incorporating correspondingly. As for performance evaluation, using data supermarket China 1 July 2020 30 June 2023, demonstrate superiority over standalone RF XGBoost, Mean Absolute Percentage Error (MAPE) reduction 10% 12%, respectively, coefficient determination (R

Язык: Английский

An improved family of unbiased ratio estimators for a population distribution function DOI Creative Commons

Sohail Ahmad,

Moiz Qureshi, Hasnain Iftikhar

и другие.

AIMS Mathematics, Год журнала: 2025, Номер 10(1), С. 1061 - 1084

Опубликована: Янв. 1, 2025

<p>This study discusses a novel family of unbiased ratio estimators using the Hartley-Ross (HR) method. The are designed to estimate population distribution function (PDF) in context simple random sampling with non-response. To assess their performance, expressions for variance obtained up initial (first) approximation order. efficiency proposed is evaluated analytically and numerically compared existing estimators. In addition, accuracy assessed four real-world datasets simulation analysis. estimator demonstrates exceptional performance under sampling, achieving percentage relative efficiencies 272.052,301.279,214.1214, 280.9528 across distinct populations, significantly outperforming For non-response different weights, exhibits remarkable efficiency, $ w_1 = 339.7875, w_2 334.6623, w_3 337.7393 Population 1, 257.0119, 274.7351, 316.0341 2, 231.8627, 223.0608, 219.9059 3, 261.3122, 242.7319, 240.0694 4, validating its robustness superiority.</p>

Язык: Английский

Процитировано

2
1VMD-ATT-LSTM Electricity Price Prediction Based on Grey Wolf Optimization Algorithm in Electricity Markets Considering Renewable Energy DOI
Yuzhen Xu, Xin Huang, Xidong Zheng

и другие.

Renewable Energy, Год журнала: 2024, Номер unknown, С. 121408 - 121408

Опубликована: Сен. 1, 2024

Язык: Английский

Процитировано

4
Transformer-based short-term traffic forecasting model considering traffic spatiotemporal correlation DOI Creative Commons

Ande Chang,

Yuting Ji, Yiming Bie

и другие.

Frontiers in Neurorobotics, Год журнала: 2025, Номер 19

Опубликована: Янв. 23, 2025

Traffic forecasting is crucial for a variety of applications, including route optimization, signal management, and travel time estimation. However, many existing prediction models struggle to accurately capture the spatiotemporal patterns in traffic data due its inherent nonlinearity, high dimensionality, complex dependencies. To address these challenges, short-term model, Trafficformer, proposed based on Transformer framework. The model first uses multilayer perceptron extract features from historical data, then enhances spatial interactions through Transformer-based encoding. By incorporating road network topology, mask filters out noise irrelevant interactions, improving accuracy. Finally, speed predicted using another perceptron. In experiments, Trafficformer evaluated Seattle Loop Detector dataset. It compared with six baseline methods, Mean Absolute Error, Percentage Root Square Error used as metrics. results show that not only has higher accuracy, but also can effectively identify key sections, great potential intelligent control optimization refined resource allocation.

Язык: Английский

Процитировано

0
A novel hybrid framework for forecasting stock indices based on the nonlinear time series models DOI
Hasnain Iftikhar, Faridoon Khan, Elías A. Torres Armas

и другие.

Computational Statistics, Год журнала: 2025, Номер unknown

Опубликована: Март 11, 2025

Язык: Английский

Процитировано

0
Forecasting day-ahead electric power prices with functional data analysis DOI Creative Commons
Faheem Jan, Hasnain Iftikhar, Muhammad Junaid Tahir

и другие.

Frontiers in Energy Research, Год журнала: 2025, Номер 13

Опубликована: Март 28, 2025

Day-ahead electricity prices in today’s competitive electric power markets have complex features such as high frequency, volatility, non-linearity, non-stationarity, mean reversion, multiple periodicities, and calendar effects. These complicated make price forecasting difficult. To address this, this research examines the application of functional data analysis to day-ahead prices. Compared classical time series approaches, is more appealing since it anticipates daily profile, allowing for short-term projections. This technique uses a autoregressive ( F AR) with exogenous predictors id="m2">X ) model predict next-day In addition, standard time-series models, including (AR) id="m4">

Язык: Английский

Процитировано

0
Electricity consumption forecasting using a novel homogeneous and heterogeneous ensemble learning DOI Creative Commons
Hasnain Iftikhar, Justyna Żywiołek, Javier Linkolk López‐Gonzales

и другие.

Frontiers in Energy Research, Год журнала: 2024, Номер 12

Опубликована: Сен. 4, 2024

In today’s world, a country’s economy is one of the most crucial foundations. However, industries’ financial operations depend on their ability to meet electricity demands. Thus, forecasting consumption vital for properly planning and managing energy resources. this context, new approach based ensemble learning has been developed predict monthly consumption. The method divides time series into deterministic stochastic components. component, which consists secular long-term trend an annual seasonality, estimated using multiple regression model. contrast, part considers short-run random fluctuations series. It forecasted by four different series, machine models, three novel proposed models: homogeneous model, heterogeneous study analyzed data Pakistan’s from 1991-January 2022-December. evaluation models criteria: accuracy metrics (including mean absolute percent error (MAPE), (MAE), root squared (RMSE), relative (RRSE)); equality forecast statistical test (the Diebold Mariano’s test); graphical assessment. model’s results show lower values compared singles with measured MAPE, MAE, RMSE, RRSE at 5.0027, 460.4800, 614.5276, 0.2933, respectively. Additionally, model statistically significant (p < 0.05) superior rest models. Also, demonstrates considerable performance least error, comparatively better than individual best reported in literature are considered baseline Further, values’ behavior depicts that higher during summer season, demand will be highest June July. graph reveal rapidly increases time. This indirectly indicates government Pakistan must take adequate steps improve production through sources restore economic status meeting demand. Despite several studies conducted various perspectives, no analysis undertaken Pakistan.

Язык: Английский

Процитировано

2
Economic determinants and forecasting of electricity demand in Cameroon: A policy-driven approach using multilinear regression DOI

Théodore Patrice Nna Nna,

Flavian Emmanuel Sapnken, Jean Gaston Tamba

и другие.

Energy 360., Год журнала: 2024, Номер unknown, С. 100013 - 100013

Опубликована: Дек. 1, 2024

Язык: Английский

Процитировано

2
Short-term PM2.5 forecasting using a unique ensemble technique for proactive environmental management initiatives DOI Creative Commons
Hasnain Iftikhar, Moiz Qureshi, Justyna Żywiołek

и другие.

Frontiers in Environmental Science, Год журнала: 2024, Номер 12

Опубликована: Сен. 10, 2024

Particulate matter with a diameter of 2.5 microns or less ( PM2.5 ) is significant type air pollution that affects human health due to its ability persist in the atmosphere and penetrate respiratory system. Accurate forecasting particulate crucial for healthcare sector any country. To achieve this, current work, new time series ensemble approach proposed based on various linear (autoregressive, simple exponential smoothing, autoregressive moving average, theta) nonlinear (nonparametric neural network autoregressive) models. Three models are also developed, each employing distinct weighting strategies: equal distribution weight among all single (ESME), assignment training average accuracy errors (ESMT), validation mean measures (ESMV). This technique was applied daily id="m3">PM2.5 concentration data from 1 January 2019, 31 May 2023, Pakistan’s main cities, including Lahore, Karachi, Peshawar, Islamabad, forecast short-term id="m4">PM2.5 concentrations. When compared other models, best model (ESMV) demonstrated ranging 3.60% 25.79% 0.81%–13.52% 1.08%–7.06% 1.09%–12.11% Peshawar. These results indicate more efficient accurate id="m5">PM2.5 than existing Furthermore, using model, made next 15 days (June June 2023). The showed highest id="m6">PM2.5 value (236.00 id="m7">μg/m3 observed 8 2023. Other displayed higher poor quality throughout days. Conversely, Karachi experienced moderate id="m8">PM2.5 levels between 50 id="m9">μg/m3 80 id="m10">μg/m3 . In id="m11">PM2.5 were consistently unhealthy, peak (153.00 id="m12">μg/m3 9 experience can assist environmental monitoring organizations implementing cost-effective planning minimize pollution.

Язык: Английский

Процитировано

1
Short‐Term Electricity Price Forecasting Using the Empirical Mode Decomposed Hilbert‐LSTM and Wavelet‐LSTM Models DOI Creative Commons
Kunal Shejul,

R Harikrishnan,

Amit Kukker

и другие.

Journal of Electrical and Computer Engineering, Год журнала: 2024, Номер 2024(1)

Опубликована: Янв. 1, 2024

The electricity sector deregulation has led to the formation of short‐term power markets where consumers can purchase by bidding at market. market price is volatile and changes are due change in demand bid different span time during day. availability forecast essential for participants make informed decisions. In this paper, modified LSTM approach, wavelet‐LSTM, Hilbert‐LSTM proposed predict objective improve precision adaptability predictions utilizing temporal dependence identification capability multiresolution analysis transforms. models combine these two effective methods order capture both long‐term trends variations present series data. 8‐year dataset used training models, based on day‐ahead calculated compared with testing techniques show better performance terms rank correlation, mean square error, root error existing algorithms CNN‐LSTM. prediction results achieved wavelet‐LSTM (1‐month 8 years) correlation 0.9746 0.9749, MSE 0.2962 0.1363, RMSE 0.5443 0.3692, respectively. than forecasting improved 61% 43%, respectively, method. Also, complete years all 12 months Hilbert‐LSTM, 0.9645, 0.3876, 0.6225. parameters conventional approaches. be accurately

Язык: Английский

Процитировано

1
Advances in time series forecasting: innovative methods and applications DOI Creative Commons
J. F. Torres, M. Martínez-Ballesteros, Alicia Troncoso

и другие.

AIMS Mathematics, Год журнала: 2024, Номер 9(9), С. 24163 - 24165

Опубликована: Янв. 1, 2024

Язык: Английский

Процитировано

0