Transforming Wind Data into Insights: A Comparative Study of Stochastic and Machine Learning Models in Wind Speed Forecasting

In this study, we benchmark various machine learning techniques against a synthetic but physically based reference time series (model-simulated (ERA5-Land/FLake) bottom-temperature series) and assess whether decomposition methods (VMD and EMD) improve forecast accuracy using Support Vector Machine (SVM), Multi-Layer Perceptron (MLP), Random Forest (RF), Gaussian Process Regression (GPR), and Long Short-Term Memory (LSTM) with the monthly average data of Mjøsa, the largest lake in Norway, between 1950 and 2024 from the ERA5-Land FLake model. A total of 70% of the dataset was used for training and 30% was reserved for testing. To assess the performance several metrics, correlation coefficient (r), Nash–Sutcliffe efficiency (NSE), Kling–Gupta efficiency (KGE), Performance Index (PI), RMSE-based RSR, and Root Mean Square Error (RMSE) were used. The results revealed that without decomposition, the GPR-M03 combination outperforms other models (with scores r = 0.9662, NSE = 0.9186, KGE = 0.8786, PI = 0.0231, RSR = 0.2848, and RMSE = 0.2000). Considering decomposition cases, when VMD is applied, the SVM-VMD-M03 combination achieved better results compared to other models (with scores r = 0.9859, NSE = 0.9717, KGE = 0.9755, PI = 0.0135, RSR = 0.1679, and RMSE = 0.1179). Conversely, with decomposition cases, when EMD applied, LSTM-EMD-M03 is explored as the more effective combination than others (with scores r = 0.9562, NSE = 0.9008, KGE = 0.9315, PI = 0.0256, RSR = 0.2978, and RMSE = 0.3143). The results demonstrate that GPR and SVM, coupled with VMD, yield high correlation (e.g., r ≈ 0.986) and low RMSE (~0.12), indicating the ability to reproduce FLake dynamics rather than as accurate predictions of measured bottom temperature.

The advancement towards utilizing renewable energy sources is crucial for mitigating environmental issues such as air pollution and climate change. Offshore wind turbines, particularly floating offshore wind turbines (FOWTs), are developed to harness the stronger, steadier winds available over deep waters. Accurate metocean data forecasts, encompassing wind speed and wave height, are crucial for offshore wind farms’ optimal placement, operation, and maintenance and contribute significantly to FOWT’s efficiency, safety, and lifespan. This study examines the application of three machine learning (ML) models, including Facebook Prophet, Seasonal Autoregressive Integrated Moving Average with Exogenous Factors (SARIMAX), and long short-term memory (LSTM), to forecast wind speeds and significant wave heights, using data from a buoy situated in the Pacific Ocean. The models are evaluated based on their ability to predict 1-, 3-, and 30-day future wind speed and wave height values, with performances assessed through Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) metrics. Among the models, LSTM displayed superior performance, effectively capturing the complex temporal dependencies in the data. Incorporating exogenous variables, such as atmospheric conditions and gust speed, further refined the predictions.The study’s findings highlight the potential of machine learning (ML) models to enhance the integration and reliability of renewable energy sources through accurate metocean forecasting.

Dengue hemorrhagic fever (DHF) is one of the most widespread and deadly diseases in several parts of Indonesia. An accurate forecast-based model is required to reduce the incidence rate of this disease. Time-series methods such as autoregressive integrated moving average (ARIMA) models are used in epidemiology as statistical tools to study and forecast DHF and other infectious diseases. The present study attempted to forecast the monthly confirmed DHF cases via a time-series approach. The ARIMA, seasonal ARIMA (SARIMA), and long short-term memory (LSTM) models were compared to select the most accurate forecasting method for the deadly disease. The data were obtained from the Surabaya Health Office covering January 2014 to December 2016. The data were partitioned into the training and testing sets. The best forecasting model was selected based on the lowest values of accuracy metrics such as the root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE). The findings demonstrated that the SARIMA (2,1,1) (1,0,0) model was able to forecast the DHF outbreaks in Surabaya City compared to the ARIMA (2,1,1) and LSTM models. We further forecasted the DHF cases for 12 month horizons starting from January 2017 to December 2017 using the SARIMA (2,1,1) (1,0,0), ARIMA (2,1,1), and LSTM models. The results revealed that the SARIMA (2,1,1) (1,0,0) model outperformed the ARIMA (2,1,1) and LSTM models based on the goodness-of-fit measure. The results showed significant seasonal outbreaks of DHF, particularly from March to September. The highest cases observed in May suggested a significant seasonal correlation between DHF and air temperature. This research is the first attempt to analyze the time-series model for DHF cases in Surabaya City and forecast future outbreaks. The findings could help policymakers and public health specialists develop efficient public health strategies to detect and control the disease, especially in the early phases of outbreaks.

The importance of day-ahead load forecasting cannot be overstated. Electricity load forecasting is highly important because it allows electric distribution utilities to increase their transmission efficiency and their revenues, increase the reliability of power supply, and correct decisions for future developments. In this paper, day-ahead load forecasting was studied using multivariate time series analysis. Traditional forecasting method such as seasonal autoregressive integrated moving average (SARIMA) was used. Machine learning algorithms such as Feed Forward Neural Network (FFNN) and Long Short Term Memory (LSTM) were also utilized to determine their applicability to short-term load forecasting. For SARIMA, only historical hourly load was needed while FFNN and LSTM required the addition of temperature data, hour of the day and day of the week, special events and previous hour load. In the prediction, FFNN and LSTM performed better with mean absolute percentage error (MAPE) of 1.80 and 1.75%, respectively, compared with SARIMA with a MAPE of 4.48%. The study demonstrated that machine learning like FFNN and LSTM outperformed the traditional SARIMA models, highlighting their effectiveness in applications such as short-term load forecasting in time series prediction.

This study evaluates the performance of traditional statistical methods, specifically the Seasonal Autoregressive Integrated Moving Average (SARIMA), in comparison to advanced deep learning models—Long Short-Term Memory (LSTM), Bidirectional LSTM (BiLSTM), and Convolutional Neural Network (CNN)—for wind speed forecasting across various regions in Saudi Arabia: Al-Jouf, Abha, Al-Ahsa, and Al-Dawadami. The historical wind speed dataset covering the year 2018 was utilized, with data preprocessing conducted using the Exploratory Data Analysis (EDA) method to ensure consistency and quality. The models were assessed based on three primary error metrics: Mean Absolute Error (MAE), Mean Squared Error (MSE), and Root Mean Square Error (RMSE). Results show that CNN consistently outperforms SARIMA and the other deep learning models, particularly when forecasting stable wind speed patterns. LSTM demonstrates an ability to handle fluctuating wind speeds effectively, while BiLSTM offers advantages in capturing complex bidirectional temporal dependencies. On the other hand, SARIMA generally exhibits lower performance compared to deep learning approaches. The superior performance of CNN is likely attributed to its strength in local feature extraction and handling spatial patterns effectively, which are beneficial for short-term forecasting. These findings provide valuable insights into model selection for wind energy forecasting and can contribute to optimizing renewable energy integration and planning strategies. Future work may explore hybrid model approaches to further enhance forecasting accuracy.

BackgroundThe early warning model of infectious diseases plays a key role in prevention and control. This study aims to using seasonal autoregressive fractionally integrated moving average (SARFIMA) model to predict the incidence of hemorrhagic fever with renal syndrome (HFRS) and comparing with seasonal autoregressive integrated moving average (SARIMA) model to evaluate its prediction effect.MethodsData on notified HFRS cases in Weifang city, Shandong Province were collected from the official website and Shandong Center for Disease Control and Prevention between January 1, 2005 and December 31, 2018. The SARFIMA model considering both the short memory and long memory was performed to fit and predict the HFRS series. Besides, we compared accuracy of fit and prediction between SARFIMA and SARIMA which was used widely in infectious diseases.ResultsModel assessments indicated that the SARFIMA model has better goodness of fit (SARFIMA (1, 0.11, 2)(1, 0, 1)12: Akaike information criterion (AIC):-631.31; SARIMA (1, 0, 2)(1, 1, 1)12: AIC: − 227.32) and better predictive ability than the SARIMA model (SARFIMA: root mean square error (RMSE):0.058; SARIMA: RMSE: 0.090).ConclusionsThe SARFIMA model produces superior forecast performance than the SARIMA model for HFRS. Hence, the SARFIMA model may help to improve the forecast of monthly HFRS incidence based on a long-range dataset.

Background: Brucellosis runs rampant endemically with sporadic outbreaks in Algeria. The present study aimed to provide insights into the epidemiology of brucellosis and compare the performance of some prediction models using surveillance data from Tebessa province, Algeria. Study Design: A retrospective study.Methods: Seasonal autoregressive integrated moving average (SARIMA), neural network autoregressive (NNAR), and hybrid SARIMA-NNAR models were developed to predict monthly brucellosis notifications. The prediction performance of these models was compared using root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE). Results: Overall, 13670 human brucellosis cases were notified in Tebessa province from 2000-2020 with a male-to-female ratio of 1.3. The most affected age group was 15-44 years (56.2%). The cases were reported throughout the year with manifest seasonality. The annual notification rate ranged from 30.9 (2013) to 246.7 (2005) per 100000 inhabitants. The disease was not evenly distributed, rather spatial and temporal variability was observed. The SARIMA (2,1,3) (1,1,1)12', NNAR (12,1,6)12' , and SARIMA (2,0,2) (1,1,0)12- NNAR (5,1,4)12 were selected as the best-fitting models. The RMSE, MAE, and MAPE of the SARIMA and SARIMA-NNAR models were by far lower than those of the NNAR model. Moreover, the SARIMA-NNNAR hybrid model achieved a slightly better prediction accuracy for 2020 than the SARIMA model. Conclusion: As evidenced by the obtained results, both SARIMA and hybrid SARIMA-NNAR models are suitable to predict human brucellosis cases with high accuracy. Reasonable predictions, along with mapping brucellosis incidence, could be of great help to veterinary and health policymakers in the development of informed, effective, and targeted policies, as well as timely interventions.

The purpose of this study is to predict the decade rainfall in Semarang City using two main methods, namely Seasonal Autoregressive Integrated Moving Average with Exogenous Variables (SARIMAX) and Long Short Term Memory (LSTM). The methodology of this study begins with data pre-processing, which includes data deletion analysis using dropna and data normalization using Min-Max Scaling to reduce scale the to between 0 and 1. The dataset is then divided into 80% training data and 20% test data. The validity of the data (X_test, Y_test) using the best 56-epoch data validation (val_loss) is better than the validity of the training data (loss). On the other hand, SARIMAX uses the (2,1,2), (2,1,2,36) model, and its validation techniques include Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Coefficient of Determination (R2). Specifically, the RMSE of the LSTM model is 19.6, and its RMSE of the SARIMAX is 31.05. The MAE of LSTM is 15.0, SARIMAX is 24.5, and the R2 of LSTM is 0.814, and SARIMAX is 0.52. Lower RMSE and MAE values indicate lower prediction errors, but a higher R2 value of 1 indicates that LSTM can explain 81% of the actual data variation, which is better than SARIMAX, which is only about 52%. The main finding of this study is that the LSTM model performs better when recommending rainfall datasets.

This study conducts a comprehensive empirical analysis of modeling and forecasting the daily stock prices of PZ Cussons Nigeria PLC using the Seasonal Autoregressive Integrated Moving Average (SARIMA) framework within the Box-Jenkins methodology. Utilizing a dataset of 2,289 daily closing prices spanning from January 5, 2015, to November 22, 2023, the research rigorously compares the forecasting performance of SARIMA against a non-seasonal ARIMA benchmark. Stationarity is assessed via the Augmented Dickey-Fuller (ADF) test, with model selection guided by the Akaike Information Criterion (AIC). The optimal models identified are SARIMA(1,0,1)(0,3,2,20) and ARIMA(5,3,0), with diagnostic checks confirming adequacy through residual analysis and Ljung-Box tests. Forecast accuracy is evaluated out-of-sample using root mean squared error (RMSE), mean absolute percentage error (MAPE), and mean absolute error (MAE) across multiple horizons. Results indicate that while ARIMA exhibits marginally lower RMSE in short-term forecasts, SARIMA consistently outperforms in MAPE and demonstrates superior robustness over medium to long-term horizons, particularly beyond one month. Trading strategy backtests further reveal that SARIMA-based signals yield significantly higher cumulative returns (15.5% vs. 8.1%), hit rates (56% vs. 47%), and Sharpe ratios (1.12 vs. 0.68), with lower maximum drawdowns. These findings underscore the critical role of seasonality potentially driven by institutional rebalancing, fiscal cycles, and behavioral market responses in Nigerian equity price dynamics. The study contributes to the literature on financial econometrics in emerging African markets by providing empirical evidence that SARIMA offers enhanced predictive accuracy and practical trading utility compared to traditional ARIMA, affirming the relevance of classical time series models in contemporary quantitative finance.

This study evaluates the effectiveness of Seasonal Autoregressive Integrated Moving Average (SARIMA) models in forecasting trends of Pediatric Diarrhea and Pneumonia in Rivers State, Nigeria, with particular emphasis on handling auto-correlated errors. The objectives include analyzing seasonal patterns, assessing autocorrelation structures, comparing alternative SARIMA specifications, and identifying the most suitable model using both real and simulated datasets. The empirical data comprise monthly records of reported cases among children under five years, obtained from the Rivers State Primary Healthcare Database. Additionally, a synthetic time series was generated to replicate similar autocorrelation characteristics for robustness testing. Autocorrelation diagnostics, including the Ljung-Box test and ACF/PACF plots, revealed significant serial dependencies across all series, underscoring the limitations of models that ignore autocorrelation. Competing SARIMA models were estimated and evaluated based on Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), Root Mean Square Error (RMSE), and residual diagnostic checks. Among the models assessed, SARIMA(0,1,2)(0,0,2)[12][12][12] emerged as the best-fitting model, particularly for pediatric diarrhea, demonstrating strong predictive performance and adherence to diagnostic assumptions. Although Auto ARIMA achieved comparable results for pneumonia, SARIMA models exhibited superior residual behaviour and diagnostic reliability. By integrating actual and simulated data, this research addresses gaps in previous studies that overlooked seasonality and autocorrelation in padiatric health time series modeling. The findings affirm the usefulness of SARIMA for time series epidemiology and highlight its potential for enhancing disease investigation and informing public health interventions in resource-constrained settings such as Rivers State.

Background: Malaria remains a significant public health problem in India, particularly in endemic states such as Karnataka. Accurate forecasting of malaria incidence is essential for strengthening surveillance systems, optimizing resource allocation and enabling timely preventive interventions. Objective: This study aimed to analyze the temporal pattern of malaria cases in Karnataka State, India and to develop an appropriate Seasonal Autoregressive Integrated Moving Average (SARIMA) model for forecasting future monthly malaria incidence. Methods: A retrospective time series analysis was conducted using monthly malaria case data from January 2020 to August 2025 obtained from the National Centre for Vector Borne Diseases Control (NCVBDC), Government of India. The Box–Jenkins methodology was applied for model development. Data preprocessing included logarithmic transformation and seasonal and non-seasonal differencing to achieve stationary. Model identification was guided by autocorrelation function (ACF) and partial autocorrelation function (PACF) plots. Competing SARIMA models were estimated using maximum likelihood estimation and evaluated using Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and Bayesian Information Criterion (BIC). Residual diagnostics were assessed using the Ljung–Box Q test. Results: The malaria time series demonstrated a clear increasing trend with pronounced seasonal variation. Among the candidate models, the SARIMA (2,1,0) (0,1,1) ₁₂ model showed the best overall performance, with the lowest RMSE (29.413), MAE (18.279), MAPE (40.437) and normalized BIC (7.054). The Ljung–Box Q test indicated no residual autocorrelation (p = 0.77). Twelve-month forecasts (September 2025–August 2026) revealed marked seasonal peaks, with forecasted values ranging from 43 to 461 cases. Conclusion: The SARIMA (2,1,0) (0,1,1) ₁₂ model effectively captured the trend and seasonal dynamics of malaria transmission in Karnataka. SARIMA based forecasting offers a practical and reliable tool for malaria surveillance and can support evidence-based planning and timely public health interventions. Keywords: Malaria, Time series analysis, SARIMA, Forecasting, Karnataka, Public health

Cow-calf operations form the foundation of the U.S. beef industry, with approximately 27.9 million beef cows and 25.1 million calves born in 2024, comprising 61.1% of total cattle inventory. Though their exact economic contribution is hard to predict due to the integrated nature of the beef supply chain, the cattle industry contributed approximately $88.4 billion (17%) to the total U.S. agricultural receipts in 2023. Despite their significance, these operations face economic challenges, including fluctuating feed costs, market volatility, and drought. Given the narrow profit margins of these operations, predictive modeling of profitability can offer valuable insights for sustainable management and decision-making. This study applies three univariate (Auto-Regressive Integrated Moving Average (ARIMA), Seasonal ARIMA (SARIMA), SARIMA with Exogenous regressors (SARIMAX)), three multivariate machine learning models (Random Forest (RF), AdaBoost, Support Vector Regression (SVR)), and one deep learning model (Long Short-Term Memory, LSTM) to assess the economic performance of cow-calf operations, specifically predicting the average farm price received for calves ($/Cwt). The analysis leverages historical data along with key domain-specific variables, including feed costs (monthly average U.S. farm price of corn, $/bushel), fuel prices (industrial natural gas price, $/thousand cubic feet), foreign exchange rates (U.S.-Japan exchange rate), and inflation indices (Consumer Price Index, CPI), to provide robust and enhanced predictive insights. A cow-calf budget dataset spanning from 1996 to 2024 was compiled from sources including the USDA Agricultural Marketing Service and Texas A&M AgriLife Extension. Data acquisition was facilitated through web scraping techniques, enabling the automated extraction of the historical economic indicators. The collected data underwent preprocessing and feature engineering, to enhance model input quality, and both univariate time-series and multivariate machine learning models were applied to capture temporal and nonlinear patterns in calf prices. Model performance was evaluated using multiple metrics, including Mean Absolute Error (MAE), Mean Percentage Error (MPE), Root Mean Square Error (RMSE), Root Mean Square Percentage Error (RMSPE), and Coefficient of Determination (R²) to comprehensively assess predictive performance. The analysis revealed substantial relationships between selling price of cow-calf operations and key economic indicators. Among the evaluated forecasting models, SARIMA exhibited the strongest performance by effectively capturing seasonal trends, while SARIMAX showed comparable accuracy through the integration of external variables. Among the machine learning techniques, AdaBoost performed best, outperforming RF and SVR. The LSTM model achieved the overall best performance, demonstrating strong capability in modeling both nonlinear dependencies and temporal dynamics in calf price trends. These findings demonstrate the potential of combining classical time-series and modern machine learning models to support price forecasting in cow-calf operations. This hybrid modeling approach provides a powerful decision-support tool for producers navigating market volatility, and future research should explore real-time forecasting and broader economic and environmental integrations to further strengthen predictive capabilities.

Objective: The article aims to explore and evaluate a novel deep-learning approach for time series forecasting using three specific models: SARIMA (Seasonal Auto-Regressive Integrated Moving Average), Neural Prophet, and Facebook Prophet. The primary goal is to assess the effectiveness of these models in predicting stock market values in the Gulf region, providing insights into the best-suited models for forecasting tasks. Methods: The study employs Python libraries and frameworks to implement the SARIMA, Neural Prophet, and Facebook Prophet models. The models are trained using stock market data from the Mulkia Gulf Real Estate dataset. The methodology includes data preprocessing, model training, evaluation, and comparison using metrics such as Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and Mean Absolute Percentage Error (MAPE). Results: The evaluation results show that SARIMA performs well in general prediction tasks, especially when datasets contain seasonality trends. Facebook Prophet excels with smaller datasets containing seasonal data, while Neural Prophet demonstrates its ability to capture complex, non-linear patterns. However, Neural Prophet requires more intricate data and fine-tuning for optimal results. Contribution: This study provides a comparative analysis of deep-learning models for time series forecasting, offering valuable insights into their strengths and weaknesses. The findings contribute to the understanding of which models are most suitable for stock market prediction and how they can be adapted to different data types and scenarios. Conclusion: The research concludes that each model—SARIMA, Facebook Prophet, and Neural Prophet—has its unique strengths in time series forecasting. SARIMA is reliable for handling seasonal data, Facebook Prophet is efficient for smaller datasets with clear trends, and Neural Prophet is best for more complex datasets. The study highlights the importance of selecting the appropriate model based on the specific requirements of the forecasting task.

Wind speed forecasting has proven to be crucial for many domains like power generation. However, it is quite convoluted and demanding due to the unpredictability of the wind. Numerous physical and statistical methods have been employed in the past by the researchers in India and over the years, these methods have been improvised. While the physical methods prove to be computationally expensive and the statistical methods represent weaker generalizations of non-linear features, there arises a need to develop much more efficient and robust approaches that can augment with the traditional methodologies. Artificial Intelligence has shown tremendous potential in providing solutions to time-series forecasting tasks, akin to the precise task that this work aims to address. Various machine learning and deep learning algorithms have been found to be a prominent choice for forecasting of various weather parameters. This work aims at implementing two machine learning algorithms, namely, Light Gradient Boosting Machine (LGBM) and Long Short-Term Memory (LSTM) network to obtain short-term forecasts of wind speed of Indian weather stations. Robust feature engineering approaches like Miss Forest to fill in the missing values, a novel feature representation for giving importance to recent data, and Fast Fourier Transform (FFT) with Digital Filters for removal of outliers and noises in the data have been used. Finally, stacking, an ensemble learning has been implemented using LGBM and LSTM as base learners and Random Forest as the metalearner. The average Root Mean Square Error (RMSE) (in m/s to tenths) obtained by individual models is high. 24-hour wind forecast errors of LGBM and LSTM are 8.19 and 21.17, whereas the forecast error using the proposed ensemble framework is 5.74.

Background: This study was designed to find the best statistical approach to scorpion sting predictions.Study Design: A retrospective study.Methods: Multiple regression, seasonal autoregressive integrated moving average (SARIMA), neural network autoregressive (NNAR), and hybrid SARIMA-NNAR models were developed to predict monthly scorpion sting cases in El Oued province. The root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE) were used to quantitatively compare different models.Results: In general, 96909 scorpion stings were recorded in El Oued province from 2005-2020. The incidence rate experienced a gradual decrease until 2012 and since then slight fluctuations have been noted. Scorpion stings occurred throughout the year with peaks in September followed by July and August and troughs in December and January. Sting cases were not evenly distributed across demographic groups; the most affected age group was 15-49 years, and males were more likely to be stung. Of the reported deaths, more than half were in children 15 and younger. Scorpion’s activity was conditioned by climate factors, and temperature had the highest effect. The SARIMA(2,0,2)(1,1,1)12, NNAR(1,1,2)12, and SARIMA(2,0,2)(1,1,1)12-NNAR(1,1,2)12 were selected as the best-fitting models. The RMSE, MAE, and MAPE of the SARIMA and SARIMA-NNAR models were lower than those of the NNAR model in fitting and forecasting; however, the NNAR model could produce better predictive accuracy.Conclusion: The NNAR model is preferred for short-term monthly scorpion sting predictions. An in-depth understanding of the epidemiologic triad of scorpionism and the development of predictive models ought to establish enlightened, informed, better-targeted, and more effective policies.