Seasonal Autoregressive Research Articles

Advanced Hybrid Models for Air Pollution Forecasting: Combining SARIMA and BiLSTM Architectures

This study explores a hybrid forecasting framework for air pollutant concentrations (PM10, PM2.5, and NO2) that integrates Seasonal Autoregressive Integrated Moving Average (SARIMA) models with Bidirectional Long Short-Term Memory (BiLSTM) networks. By leveraging SARIMA’s strength in linear and seasonal trend modeling and addressing nonlinear dependencies using BiLSTM, the framework incorporates Box-Cox transformations and Fourier terms to enhance variance stabilization and seasonal representation. Additionally, attention mechanisms are employed to prioritize temporal features, refining forecast accuracy. Using five years of daily pollutant data from Romania’s National Air Quality Monitoring Network, the models were rigorously evaluated across short-term (1-day), medium-term (7-day), and long-term (30-day) horizons. Metrics such as RMSE, MAE, and MAPE revealed the hybrid models’ superior performance in capturing complex pollutant dynamics, particularly for PM2.5 and PM10. The SARIMA combined with BiLSTM, Fourier, and Attention configuration demonstrated consistent improvements in predictive accuracy and interpretability, with attention mechanisms proving effective for extreme values and long-term dependencies. This study highlights the benefits of combining statistical preprocessing with advanced neural architectures, offering a robust and scalable solution for air quality forecasting. The findings provide valuable insights for environmental policymakers and urban planners, emphasizing the potential of hybrid models for improving air quality management and decision-making in dynamic urban environments.

A Model-Data Dual-Driven Approach for Predicting Shared Bike Flow near Metro Stations

Bike-sharing has emerged as an innovative green transportation mode, showing promising potential in addressing the ‘last-mile’ transportation challenge in an eco-friendly manner. However, shared bikes around metro stations often face supply–demand imbalance problems during peak hours, causing bike shortages or congestion that compromise user experience and bike utilization. Accurate prediction enables operators to develop rational dispatch strategies, improve bike turnover rate, and promote synergistic metro–bike integration. However, state-of-the-art research predominantly focuses on improving complex deep-learning models while overlooking their inherent drawbacks, such as overfitting and poor interpretability. This study proposes a model–data dual-driven approach that integrates the classical statistical regression model as a model-driven component and the advanced deep-learning model as a data-driven component. The model-driven component uses the Seasonal Autoregressive Integrated Moving Average (SARIMA) model to extract periodic patterns and seasonal variations of historical data, while the data-driven component employs an Extended Long Short-Term Memory (xLSTM) neural network to process nonlinear relationships and unexpected variations. The fusion model achieved R-squared values of 0.9928 and 0.9770 for morning access and evening egress flows, respectively, and reached 0.9535 and 0.9560 for morning egress and evening access flows. The xLSTM model demonstrates an 8% improvement in R2 compared to the conventional LSTM model in the morning egress flow scenario. For the morning egress and evening access flows, which exhibit relatively high variability, classical statistical models show limited effectiveness (SARIMA’s R2 values are 0.8847 and 0.9333, respectively). Even in scenarios like morning access and evening egress, where classical statistical models perform well, our proposed fusion model still demonstrates enhanced performance. Therefore, the proposed data–model dual-driven architecture provides a reliable data foundation for shared bike rebalancing and shows potential for addressing the challenges of limited robustness in statistical regression models and the susceptibility of deep-learning models to overfitting, ultimately enhancing transportation ecosystem sustainability.

Is There a High Rice Price Volatility in the Philippines: Insights from Bayesian Seasonal Autoregressive Distributed Lag (BARDL)

This paper analyzes the high volatility in rice prices in the Philippine market from October 1994 to April 2024 using a Bayesian seasonal autoregressive distributed lag (BARDL) model with three lags. Using a dataset of 160 observations, this model controls for lagged prices and seasonal effects that showed significant persistence in rice prices through the first lag, market corrections through the second lag, and minimal impacts from seasonality. Model diagnostics, including WAIC and LOOIC, indicate the model's goodness but show considerable unseen variability σ, and external factors seem to be prevailing. The findings thereby emphasize the failure of using market mechanisms to stabilize prices and the crucial need for strategic government interferences. It discusses market failure effects, the pendulum approach to economic policy, and welfare economics, urging for a balanced approach that combines market intervention with more supportive policies toward farmers to stabilize the rice market.

A Comprehensive Approach to CO2 Emissions Analysis in High-Human-Development-Index Countries Using Statistical and Time Series Approaches

Reducing carbon dioxide (CO2) emissions is vital at both global and national levels, given their significant role in exacerbating climate change. CO2 emissions, stemming from a variety of industrial and economic activities, are major contributors to the greenhouse effect and global warming, posing substantial obstacles in addressing climate issues. It is imperative to forecast CO2 emissions trends and classify countries based on their emission patterns to effectively mitigate worldwide carbon emissions. This paper presents an in-depth comparative study on the determinants of CO2 emissions in twenty countries with high Human Development Index (HDI), exploring factors related to economy, environment, energy use, and renewable resources over a span of 25 years. The study unfolds in two distinct phases: initially, statistical techniques such as Ordinary Least Squares (OLS), fixed effects, and random effects models are applied to pinpoint significant determinants of CO2 emissions. Following this, the study leverages supervised and unsupervised time series approaches to further scrutinize and understand the factors influencing CO2 emissions. Seasonal AutoRegressive Integrated Moving Average with eXogenous variables (SARIMAX), a statistical time series forecasting model, is first used to predict emission trends from historical data, offering practical insights for policy formulation. Subsequently, Dynamic Time Warping (DTW), an unsupervised time series clustering approach, is used to group countries by similar emission patterns. The dual-phase approach utilized in this study significantly improves the accuracy of CO2 emissions predictions while also providing a deeper insight into global emission trends. By adopting this thorough analytical framework, nations can develop more focused and effective carbon reduction policies, playing a vital role in the global initiative to combat climate change.

COMPARING FORECASTS OF AGRICULTURAL SECTOR EXPORT VALUES USING SARIMA AND LONG SHORT-TERM MEMORY MODELS

Indonesia's agricultural sector plays a crucial role in the national economy, with significant export potential and supporting the livelihoods of the majority of the population. As part of the government's vision to make Indonesia the world's food barn by 2045, increasing the volume and value of agricultural product exports is a primary focus, making export value forecasting essential for supporting strategic decision-making. Sequential data analysis is an important approach in analyzing data collected over a specific period. This study aims to compare two popular methods in forecasting the export value of the agricultural sector, namely the Seasonal AutoRegressive Integrated Moving Average (SARIMA) model and the Long Short-Term Memory (LSTM) model. Monthly agricultural export data from West Java Province from January 2013 to February 2024 were used as the dataset. The best SARIMA model generated was (1,1,1)(0,1,1)12, while the optimal parameters for the LSTM model were neuron: 50, dropout rate: 0.3, number of layers: 2, activation function: relu, epochs: 500, batch size: 64, optimizer: Adam, and learning rate: 0.01. Evaluation was performed using the Root Mean Squared Error (RMSE) method to measure the accuracy of both models in forecasting the export value of the agricultural sector. The results showed that the LSTM model outperformed the SARIMA model, with a lower RMSE value (SARIMA: 4182.133 and LSTM: 1939.02). This research provides valuable insights for decision-makers in planning agricultural sector export strategies in the future. With this comparison, it is expected to provide better guidance in selecting forecasting methods suitable for the characteristics of the data.

Seasonality of the incidence of bronchiolitis in infants - Brazil, 2016-2022: An interrupted time-series analysis.

To evaluate the seasonality of acute bronchiolitis in Brazil during the 2020-2022 season and compare it with the previous seasons. Data from the incidence of hospitalizations due to acute bronchiolitis in infants <1 year of age were obtained from the Department of Informatics of the Brazilian Public Health database for the period between 2016 and 2022. These data were also analyzed by macro-regions of Brazil (North, Northeast, Southeast, South, and Midwest). To describe seasonal and trend characteristics over time, we used the Seasonal Autoregressive Integrated Moving Averages Model. Compared to the pre-COVID-19 period, the incidence of hospitalizations related to acute bronchiolitis decreased by 97% during non-pharmacological interventions (March 2020 - August 2021) but increased by 95% after non-pharmacological interventions relaxation (September 2021 - December 2022), resulting in a 16% overall increase. During the pre-COVID-19 period, hospitalizations for acute bronchiolitis followed a seasonal pattern, which was disrupted in 2020-2021 but recovered in 2022, with a peak occurring in May, approximately 4% higher than the pre-COVID-19 peak. This study underscores the significant influence of COVID-19 interventions on acute bronchiolitis hospitalizations in Brazil. The restoration of a seasonal pattern in 2022 highlights the interplay between public health measures and respiratory illness dynamics in young children.

PERBANDINGAN METODE SARIMA DAN EXPONENTIAL SMOOTHING DALAM MEMPREDIKSI CURAH HUJAN DI KABUPATEN BANGKA TENGAH

Indonesia is a country that has a tropical climate because it is located on the equator. The tropical climate has two seasons, namely the dry season and the rainy season. Each region has a different climate intensity. Rainfall is one of the elements of weather and climate that has an influence on tropical areas such as Indonesia. To find out changes in rainfall patterns, predictions of the amount of rainfall are carried out using past data. In this research, rainfall prediction uses two methods, namely the Seasonal Auto Regressive Integrated Moving Average (SARIMA) method and exponential smoothing with rainfall data from 2018-2023. SARIMA is an ARIMA method that was developed to use data patterns that repeat seasonally. in a fixed time such as quarterly, semi-annually and annually. Exponential smoothing is a moving average forecasting technique that weighs past data exponentially so that the most recent data has a weight or greater in the moving average. Based on the results of the research, it was found that the best method between SARIMA and Exponential Smoothing Winters Additive was Exponential Smoothing Winters Additive with an RMSE value of 3.283.

Deep learning hybrid models for effective supply chain risk management: mitigating uncertainty while enhancing demand prediction

In today's rapidly evolving business landscape, effective supply chain management (SCM) is crucial for achieving success. Accurately predicting product demand is a significant challenge for companies, impacting customer satisfaction, inventory optimization, cost reduction, and operational efficiency. This study focuses on demand forecasting within intelligent supply chains (SCs) and supply chain risk management (SCRM), aiming to enhance overall SC efficiency and mitigate risks, highlighting the use of deep learning hybrid and singles models to address SCRM challenges, specifically in mitigating uncertainty and improving demand prediction accuracy. Our research paper investigates predictive modeling techniques for demand forecasting within the automotive sector. Specifically, we assess the effectiveness of Seasonal Autoregressive Integrated Moving Average (SARIMA), Long Short-Term Memory (LSTM), Artificial Neural Network (ANN), Recurrent Neural Network (RNN), and a hybrid RNN-ANN model with Gradient Boosting (GB). Through meticulous analysis and evaluation, we demonstrate the superior predictive accuracy of the hybrid model compared to individual models. The results indicate consistent outperformance of the hybrid model, as evidenced by lower Mean Absolute Error (MAE) and Mean Squared Error (MSE) values across electric and thermal product categories. This research aims to provide valuable insights and practical tools for businesses to refine their demand prediction processes. By addressing demand uncertainty, organizations can streamline their SCs, minimize costs, and establish a responsive and adaptable framework for sustainable growth.

Time series models for the description and forecasting of incident cases of dengue: a case study in Londrina- Paraná

Objective: The objective of this study is to describe the temporal behavior of dengue cases and to model the temporal evolution of reported incident cases for the Londrina city, Paraná-Brazil using time series modeling, specifically Seasonal Autoregressive Integrated Moving Average (SARIMA) models from January 2015 to February 2023, with the aim of predicting new cases. Methods: This is an ecological epidemiological study with a time series component. Data were obtained monthly by InfoDengue site in a structured way for the Londrina city- Paraná between January 2015 and February 2023. Time series analysis was used and the SARIMA (Seasonal Autoregressive Integrated Moving Average) model was chosen, since these models incorporate a seasonal component, to fit the natural situation of the disease with recursive periodic peaks and drops over the series, considering thus the intrinsic seasonality of dengue in spring and summer periods with an increase in cases. Results: After several inductive adjustments, the SARIMA model (2,1,2)(1,1,1)12 provided the best fit for dengue incidence data, with good performance, including capturing the oscillation peaks of the year 2020 (period of the coronavirus pandemic). Conclusion: The main contribution of the work is a better understanding of the disease, allowing predictions of the number of cases in periods subsequent to the series studied, which brings support for decision-making in public policies for prevention and educational actions for the population, and possibility of managing the SUS priority health system.

Research on the Training of Broadcasting and Hosting Talents in Colleges and Universities Based on SARIMA-BP Prediction Model

In order to improve the employment adaptability of students majoring in broadcasting and hosting in colleges and universities, a prediction model is constructed by combining intelligent Seasonal AutoRegressive Integrated Moving Average (SARIMA) and Back Propagation Neural network (BPNN) in this study. This paper first analyzes the crisis faced by broadcasting professionals, and on this basis, SARIMA-BPNN model to analyze the demand of broadcasting professionals, taking Root Mean Squared Error as the evaluation index. The RMSE index value is 5.012, which shows that the SARIMA-BPNN model has a good prediction effect. Finally, this study predicts the changing trend of social demand for broadcasting talents, and then puts forward suggestions for the training of broadcasting host talents in colleges and universities to help promote the development of the training of broadcasting host talents in colleges and universities.

MODELING AND FORECASTING OF TOURISM INCOME: THE CASE OF TURKEY

This study aims to identify the optimal forecasting model for predicting Türkiye's tourism income, a crucial factor for economic planning and development. This study employs different forecasting techniques, including the seasonal Autoregressive Integrated Moving Average (SARIMA), the additive and multiplicative Holt-Winters methods, the Exponential Smoothing State Space (ETS), Artificial Neural Networks (ANNs) and seasonal-trend decomposition procedure based on the loess (STL)-ANN hybrid model and evaluates their performance. The methodology involves analyzing monthly tourism income data from January 2012 to December 2023, incorporating additional economic indicators such as the economic confidence index, number of visitors, consumer price index, industrial production index, and USD exchange rate, which serve as input for ANN models. The findings reveal that ANNs, particularly the model that incorporates tourism income alongside other economic indicators, outperform traditional models with the lowest Mean Absolute Scaled Error (MASE) and Root Mean Squared Scaled Error (RMSSE). Specifically, the ANN model with additional predictors demonstrates the highest forecasting accuracy. These results suggest that advanced machine learning techniques provide superior predictive capabilities compared to traditional linear models. The study underscores the importance of integrating complex models to achieve more accurate forecasting, offering valuable insights for policymakers and practitioners in the tourism sector.

Prescriptive temporal modeling approach using climate variables to forecast dengue incidence in Córdoba, Colombia.

We present a modeling strategy to forecast the incidence rate of dengue in the department of Córdoba, Colombia, thereby considering the effect of climate variables. A Seasonal Autoregressive Integrated Moving Average model with exogenous variables (SARIMAX) model is fitted under a cross-validation approach, and we examine the effect of the exogenous variables on the performance of the model. This study uses data of dengue cases, precipitation, and relative humidity reported from years 2007 to 2021. We consider three configurations of sizes training set-test set: 182-13,189-6, and 192-3. The results support the theory of the relationship between precipitation, relative humidity, and dengue incidence rate. We find that the performance of the models improves when the time series models are previously adjusted for each of the exogenous variables, and their forecasts are used to determine the future values of the dengue incidence rate. Additionally, we find that the configurations 189-6 and 192-3 present the most consistent results with regard to the model's performance in the training and test data sets.

Comparative Time Series Analysis of SARIMA, LSTM, and GRU Models for Global SF6 Emission Management System

To maintain the balance of the atmosphere, the amount of change in greenhouse gas emissions must be under control. In order to create a management system and take forward-looking steps in this regard, there is no concrete data other than prediction models today. The success of prediction methods is better understood by comparing multiple methods. This research estimates the changes in the emissions of Sulfur Hexafluoride gas (SF6) in the atmosphere using Seasonal Autoregressive Integrated Moving Average Model (SARIMA), Long-Short Term Memory Neural Network (LSTM) and Gated Recurrent Unit (GRU) forecast models and compares their accuracies. Focusing on monthly SF6 emission values between 1998; 2023, time series analysis was performed to predict future emission figures. The actual values and forecast results were compared and evaluated with performance criteria such as R2, RMSE, NSE, MAE and MAPE%. The findings of this research highlight a continious upward trend in SF6 emissions and project that emission levels could approximately double from current levels by 2050. During the analysis process, all three methods performed well in estimating global SF6 gas emissions. The LSTM model generally outperformed SARIMA and GRU models, having the lowest MAPE (0.003%), MAE (0.0003), RMSE (0.0003), and R2 (1) values. It also exhibited very high predictive success with an NSE value of 0.9991. Therefore, it was determined to be the most suitable estimation method with the least error. The aim of this study is to contribute scientifically to the reduction strategies of SF6 emissions.

Dengue in Tomorrow: Predictive Insights From ARIMA and SARIMA Models in Bangladesh: A Time Series Analysis.

Dengue fever has been a continued public health problem in Bangladesh, with a recent surge in cases. The aim of this study was to train ARIMA and SARIMA models for time series analysis on the monthly prevalence of dengue in Bangladesh and to use these models to forecast the dengue prevalence for the next 12 months. This secondary data-based study utilizes AutoRegressive Integrated Moving Average (ARIMA) and Seasonal AutoRegressive Integrated Moving Average (SARIMA) models to forecast dengue prevalence in Bangladesh. Data was sourced from the Institute of Epidemiology Disease Control and Research (IEDCR) and the Directorate General of Health Services (DGHS). STROBE Guideline for observational studies was followed for reporting this study. The ARIMA (1,1,1) and SARIMA (1,1,2) models were identified as the best-performing models. The forecasts indicate a steady dengue prevalence for 2024 according to ARIMA, while SARIMA predicts significant fluctuations. It was observed that ARIMA (1,1,1) and SARIMA (1,2,2) (1,1,2) 12 were the most suitable models for prediction of dengue prevalence. These models offer valuable insights for healthcare planning and resource allocation, although external factors and complex interactions must be considered. Dengue prevalence is expected to rise in future in Bangladesh.

TIME SERIES, MACHINE LEARNING AND ENSEMBLE MODELS TO FORECAST RICE CROP YIELD IN NIGERIA UNDER CLIMATIC VARIABILITY

The rising demand for food globally due to unprecedented population growth has led to food insecurity in some populated regions, such as Africa. This leads to global agricultural production to be of increasing concern. The growing requirement for food in an era of a changing climate and scarce resources has inspired substantial crop yield prediction research. Another contributing factor to global food insecurity is climate change and its variability. The impact of increasing climate variability on rice crop yield is now evident. Predicting the potential effects of climate change on rice crops prompts the use of time series, machine learning, and ensemble models. The ensemble model was designed by assigning weights to the individual models based on their performance metrics, such as R-squared values and error rates. Results showed that the ensemble model achieved superior accuracy, with an R-squared value of 0.9523, surpassing the performance of the individual SARIMAX (Seasonal AutoRegressive Integrated Moving Average with eXogenous Regressors) and LSTM (Long Short-Term Memory)models. The predicted yields aligned closely with observed values in both training and testing phases, demonstrating the model's ability to capture historical trends and generalise effectively to unseen data. This study shows the importance of integrating time-series forecasting and machine learning techniques to improve rice crop yield predictions.

Research on Forecasting Sales of Pure Electric Vehicles in China Based on the Seasonal Autoregressive Integrated Moving Average–Gray Relational Analysis–Support Vector Regression Model

Aiming to address the complexity and challenges of predicting pure electric vehicle (EV) sales, this paper integrates a time series model, support vector machine and combined model to forecast EV sales in China. Firstly, a seasonal autoregressive integrated moving average (SARIMA) model was constructed using historical EV sales data, and the model was trained on sales statistics to obtain forecasting results. Secondly, variables that were highly correlated with sales were analyzed using gray relational analysis (GRA) and utilized as input parameters for the support vector regression (SVR) model, which was constructed to optimize sales predictions for EVs. Finally, a combined model incorporating different algorithms was verified against market sales data to explore the optimal sales prediction approach. The results indicate that the SARIMA-GRA-SVR model with the squared prediction error and inverse method achieved the best predictive performance, with MAPE, MAE and RMSE values of 12%, 1.45 and 2.08, respectively. This empirical study validates the effectiveness and superiority of the SARIMA-GRA-SVR model in forecasting EV sales.

A novel tucker decomposition driven taxi travel demand forecasting algorithm

&lt;p&gt;This research aims to enhance taxi travel demand forecasting for sustainable urban traffic management and planning. We extend the Seasonal Autoregressive Integrated Moving Average model into a high-dimensional tensor form by treating the urban transport network as a Euclidean space and introducing Tucker decomposition. This novel approach, both theoretically significant and practically applicable, represents time series data as tensors to better capture multimodal structures and correlations, improving predictive accuracy. Tucker decomposition reduces computational complexity and memory requirements, making it ideal for large-scale urban traffic network prediction. Experiments on six real-world datasets show that the model’s MAE and RMSE are reduced by about 39.43% and 27.01% on average, respectively, compared to the baseline model. Notably, the model is computationally very efficient and takes only a relatively short time to train., suitable for real-time traffic management, congestion mitigation, and resource optimization. In summary, this work innovates time series analysis, providing an efficient and precise tool for urban traffic management and planning, contributing to sustainable urban transportation advancement.&lt;strong&gt;&lt;/strong&gt;&lt;/p&gt;

Enhancing periodic mobility planning: ARIMA-driven prognostications for subway passenger volume dynamics

Abstract. This study aims to accurately predict urban subway passenger flow using the Auto-Regressive Integrated Moving Average (ARIMA) and Seasonal Auto Regressive Integrated Moving Average (SARIMA) model, with the intention of providing a scientific basis for subway operation management. By meticulously sorting and analyzing the existing traffic data of a large city's subway system, this research constructs a model that comprehensively considering the seasonal, trend, and stochastic characteristics of time series data. During the model construction process, data preprocessing is carried out, including stationarity tests, differencing, model order determination, and parameter estimation, ultimately identifying an ARIMA and SARIMA model with good fitting results, effectively identifying and forecasting characteristics such as numerical fluctuations. Additionally, this study conducts a special analysis of the prediction effects at diverse time scales, verifying the applicability and superiority of the ARIMA and SARIMA model in subway passenger flow prediction. This research not only provides a powerful tool for passenger flow prediction for subway operation management departments, aiding in the reasonable allocation of capacity and optimization of operation strategies, but also holds significant reference and application value for the passenger flow prediction and management of other urban public transportation systems.

Comparative Study on Forecast Performance from Decomposition, Winter’s and Sarima Models

This study investigates the forecasting accuracy of three univariate time series models—Decomposition, Winter’s method, and Seasonal Autoregressive Integrated Moving Average (SARIMA) to predict the Agricultural GDP of Nigeria. Quarterly data on Nigeria's Agricultural GDP from 2010 to the first quarter of 2023, obtained from the National Bureau of Statistics, were analyzed. The study applied Box-Jenkins SARIMA modeling, time series decomposition, and Winter’s method to compare their forecasting accuracy using Root Mean Square Error (RMSE) as the selection criterion. The results revealed that the SARIMA (0, 0, 2)(2, 1, 0) model outperformed the other methods, with the lowest RMSE, indicating its superior accuracy in forecasting Agricultural GDP. Winter’s method and the Decomposition method. The forecast from the SARIMA model indicated a positive trend in Nigeria’s Agricultural GDP over the study period, reinforcing the sector’s critical role in economic growth.

A Comprehensive Analysis of Santiago's Air Quality Forecasts Using SARIMA Model

Abstract. This study analyses Santiagos air quality by forecasting key pollutant concentrations- PM2.5, PM10, NO, and O-over June 2024 to May 2025 using the Seasonal Autoregressive Integrated Moving Average (SARIMA) model. Historical data from 2016 to 2024 shows seasonal peaks, particularly in winter, driven by residential heating and atmospheric conditions. A notable decrease in annual peak NO level, from 71 g/m in 2019 to 23 g/m in 2020, likely due to the COVID-19 pandemic and air quality initiatives. The SARIMA forecasts indicate that PM2.5 levels could reach up to 80 g/m, NO around 28 g/m, O around 47ppb and PM10 around 45g/m during peak periods, highlighting ongoing seasonal pollution challenges. To address these, policymakers should enforce stricter emissions regulations, promote cleaner heating, expand public transportation, and improve air quality monitoring. Additional strategies like traffic management, urban greening, and preparedness for high-pollution days are also recommended. This study demonstrates the SARIMA models effectiveness in forecasting air quality trends, offering valuable insights for policy development to protect public health in Santiago.