Traditional Time Series Models Research Articles

Temporal cross‐validation in forecasting: A case study of COVID‐19 incidence using wastewater data

AbstractTwo predominant methodologies in forecasting temporal processes include traditional time series models and machine learning methods. This paper investigates the impact of time series cross‐validation (TSCV) on both approaches in the context of a case study predicting the incidence of COVID‐19 based on wastewater data. The TSCV framework outlined in the paper begins by engineering interpretable features hypothesized as potential predictors of COVID‐19 incidence. Feature selection and hyperparameter tuning are then utilized with TSCV to identify the best features and hyperparameters for optimal model performance given a specific forecast horizon. While evidence supporting the utility of TSCV for auto‐regressive integrated moving average model with exogenous variables (TS‐ARIMAX) forecasts is lacking in this study, such an approach proves advantageous for gradient boosting machine forecasts (TS‐GBM). In Wyoming, for instance, TS‐GBM had a 34.9% improvement compared to naïve predictions, whereas GBM without TSCV only had a 15.6% improvement. However, TSCV also enhances interpretability for both TS‐ARIMAX and TS‐GBM models as this approach selects specific features, such as lagged values of COVID‐19 cases, based on forecast performance and forecast length. Future research should work to explore the influence of stationarity and model averaging on the performance of TSCV in forecasting applications.

Energy Market Price Forecasting and Financial Technology Risk Management Based on Generative AI

Abstract: The volatility of global energy markets, particularly electricity prices, plays a crucial role in influencing international economic activities. With the ongoing global energy transition and the push for low-carbon development, predicting electricity prices has become increasingly important for policymakers and market participants. This paper explores the forecasting capabilities of the ARIMA and LSTM models in analyzing electricity prices in the United States, drawing from data spanning 2001 to 2024.ARIMA, a traditional time series model, is valued for its simplicity and effectiveness in capturing linear trends, while LSTM, a deep learning-based model, excels at handling long-term dependencies in complex datasets. This study reveals that while both models offer valuable insights, each exhibits limitations. ARIMA struggles with non-linear patterns and volatility, whereas LSTM tends to underestimate extreme price values. The findings highlight the potential of hybrid models that combine traditional and machine learning approaches to enhance forecasting accuracy in the increasingly dynamic energy market. This research provides essential guidance for improving decision-making processes in the context of the global shift towards clean energy.

Exploiting News Analytics for Volatility Forecasting

ABSTRACTThis study investigates the potential of news sentiment in predicting stock market volatility. We augment traditional time series models of realized volatility with the sentiment of macroeconomic and firm‐specific news. Our results demonstrate that incorporating the sentiment of domestic macroeconomic news significantly improves volatility predictions for individual stocks and the S&P 500 Index. Notably, we find substantial enhancements in long‐horizon volatility predictions when including the sentiment of macroeconomic news in the regression models. In contrast, firm‐specific news sentiment shows only modest predictive power in the general framework. However, expanding the set of predictors to include the news count of firm‐specific news occurring overnight between two consecutive trading periods significantly improves one‐period‐ahead volatility forecasts.JEL Classification: C53, C55, C58, G14, G17

TrafficPro: A road speed prediction framework for urban signal-controlled road networks

In view of the problem that traditional deep learning models do not consider the active time-varying characteristics of traffic flow (signal control information) when predicting the speed of urban road networks, and have low prediction accuracy, this paper proposes a speed prediction framework based on generative adversarial networks and graph neural networks. In this framework, the generator network simultaneously encodes the road network traffic flow and signal control information through active and passive prediction modules to generate prediction results, and then uses the discriminator network to improve the generalization of the prediction results. This framework can obtain higher prediction accuracy than traditional time series models and deep learning models. In the real road network speed prediction scenario, this framework can reduce the prediction error compared to the best benchmark model 3%-4%.

Temporal Convolutional Network for Carbon Tax Projection: A Data-Driven Approach

This study introduces a novel application of a temporal convolutional network (TCN) for projecting carbon tax prices, addressing the critical need for accurate forecasting in climate policy. Utilizing data from the World Carbon Pricing Database, we demonstrate that the TCN significantly outperformed traditional time series models in capturing the complex dynamics of carbon pricing. Our model achieved a 31.4% improvement in mean absolute error over ARIMA baselines, with an MAE of 2.43 compared to 3.54 for ARIMA. The TCN model also showed superior performance across different time horizons, demonstrating a 30.0% lower MAE for 1-year projections, and enhanced adaptability to policy changes, with only a 39.8% increase in prediction error after major shifts, compared to ARIMA’s 95.6%. These results underscore the potential of deep learning for enhancing the precision of carbon price projections, thereby supporting more informed and effective climate policy decisions. Our findings have significant implications for policymakers and stakeholders in the realm of carbon pricing and climate change mitigation strategies, offering a powerful tool for navigating the complex landscape of environmental economics.

Forecasting Non-stationary Time Series Using Deep Learning in a Fuzzy Time Series Framework and its Application to Stock Markets

Non-stationary time series prediction is challenging due to its dynamic and complex nature. Fuzzy time series models offer a promising solution for forecasting such data, but a key challenge lies in partitioning the universe of discourse, which significantly impacts forecasting accuracy. Traditional fuzzy time series models often use equal-length interval partitioning, which is more suited for stationary data and limits their adaptability to non-stationary time series. This paper introduces a novel variable-length interval partitioning method designed specifically for non-stationary time series. The developed method combines a Long Short-Term Memory (LSTM) Autoencoder with K-means clustering, enabling dynamic, data-driven partitioning that adapts to the changing characteristics of the data. The LSTM Autoencoder encodes the time series, which is clustered using K-means, and intervals are defined based on cluster centers. Furthermore, the Variable Length Interval Partitioning-based Fuzzy Time Series model (VLIFTS) is developed by incorporating this partitioning method and the concepts of Markov chain and transition probability matrix. In this model, fuzzy sets are viewed as states of a Markov chain, and transition probabilities are used in the forecasting phase. The model is validated on stock market indices Nifty 50, NASDAQ, S&P 500, and Dow Jones. Stationarity and heteroscedasticity are tested using Augmented Dickey-Fuller (ADF) and Levene's tests respectively. Statistical forecast accuracy metrics Root Mean Squared Error (RMSE) and Mean Absolute Percent Error (MAPE) show that VLIFTS significantly improves forecasting accuracy over traditional models. This hybrid approach enhances fuzzy time series modelling and can be applied to various non-stationary time series forecasting problems.

Application of Time-Series Modeling in Forecasting the Doctorate-Level Science and Technology Workforce

The science and technology (S&T) workforce plays a crucial role in social development by promoting technological innovation and economic growth, as well as serving as a key indicator of research and development productivity and measure of innovation capability. Therefore, effective S&T workforce policies must be established to enhance national competitiveness. This study proposes a time-series forecasting methodology to predict the scale and structural trends of South Korea’s doctorate-level S&T workforce. Based on earlier research and case data, we applied both the traditional time-series model exponential smoothing and the latest model Prophet, developed by Meta, in this study. Further, public data from South Korea were used to apply the proposed models. To ensure robust model evaluation, we considered multiple metrics. With respect to both forecasting accuracy and sensitivity to data variability, Prophet was found to be the most suitable for predicting the S&T doctorate workforce’s scale. The scenarios derived from the Prophet model can help the government formulate policies based on scientific evidence in the future.

VWFTS-PSO: a novel method for time series forecasting using variational weighted fuzzy time series and particle swarm optimization

Variational weighted fuzzy time series models are known to perform better than traditional fuzzy time series models, due to their ability to handle outliers and noise. However, the performance of variational weighted fuzzy time series models is highly dependent on the partitioning of the universe of discourse. In this paper, we propose a novel variational weighted fuzzy model integrated with particle swarm optimization for choosing the best possible partitioning of the universe of discourse. Our approach combines the strengths of variational weighted fuzzy time series to handle uncertain and imprecise data and particle swarm optimization techniques to find the best model parameters. We evaluate the performance of the proposed approach on standard time series datasets and compare it with other forecasting methods in the literature. Our findings demonstrate that the proposed approach achieves accurate and reliable forecasts, outperforming other state-of-the-art approaches.

Relative position matrix and multi-scale feature fusion for writer-independent online signature verification

Online signature verification (OSV) is widely used in finance, law and other fields, and is one of the important research projects on biological characteristics. However, its data set has a small scale and has high requirements for generalization of certification models. Therefore, how to overcome these problems is of great value to improve the practicality and security of online handwriting signature technology. We propose a writer-independent online handwritten signature verification method, which adopts the relative position matrix method to convert the traditional temporal features into images for processing. This method enriched the features of the signatures, serving the purpose of data augmentation. Then two-dimensional multi-scale feature fusion based Siamese neural network (2D-MFFnet) is built for representing and learning the importance of each channel adaptively combined with the attention mechanism. Finally, a temporal convolutional network is designed to construct the classifier. The results illustrate that compared with traditional time series models, the algorithm has reduced the equal error rate by at least 2.52 % on the open datasets MCYT-100 and SVC2004 task2.

Predicting GDP with Machine Learning Technique

This paper proposes a method for reducing model errors in regressions when modelling macroeconomic variables by using machine learning algorithms and traditional time series regression models. In this paper, machine learning models are subjected to repeated k-fold cross validation and hyperparameter tuning. The linear model uses repeated k-fold cross validation, on the other hand, the traditional time series model Mixed Data Sampling Auto Regressive Distribution Lag model is run without repeated k-fold cross validation and hyperparameter tuning. The results show that integrating repeated k-fold cross validation with hyperparameter tuning increases the overall performance of machine learning algorithms and each model records the average outcome from all folds and runs. These findings demonstrate how machine learning models outperform the traditional time series model.

Research into Prediction Method for Pressure Pulsations in a Centrifugal Pump Based on Variational Mode Decomposition-Particle Swarm Optimization and Hybrid Deep Learning Models.

Centrifugal pump pressure pulsation contains various signals in different frequency domains, which interact and superimpose on each other, resulting in characteristics such as intermittency, non-stationarity, and complexity. Computational Fluid Dynamics (CFD) and traditional time series models are unable to handle nonlinear and non-smooth problems, resulting in low accuracy in the prediction of pressure fluctuations. Therefore, this study proposes a new method for predicting pressure fluctuations. The pressure pulsation signals at the inlet of the centrifugal pump are processed using Variational Mode Decomposition-Particle Swarm Optimization (VMD-PSO), and the signal is predicted by Convolutional Neural Networks-Long Short-Term Memory (CNN-LSTM) model. The results indicate that the proposed prediction model combining VMD-PSO with four neural networks outperforms the single neural network prediction model in terms of prediction accuracy. Relatively high accuracy is achieved by the VMD-PSO-CNN-LSTM model for multiple forward prediction steps, particularly for a forward prediction step of 1 (Pre = 1), with a root mean square error of 0.03145 and an average absolute percentage error of 1.007%. This study provides a scientific basis for the intelligent operation of centrifugal pumps.

Heterogeneous technology-induced global CO2 emission reduction and emission forecasting since the Kyoto era

While identifying the drivers of global CO2 emission is crucial for climate change mitigation, the heterogeneous technology-related drivers (e.g., technological change and efficiency of energy and CO2 emission) were ignored to a large extent at a global scale, hindering to formulate heterogenous climate policies. Moreover, the projection of CO2 emission was also not well compared for countries. Here, the study investigated the heterogeneous technology-related drivers of CO2 emissions in time and space simultaneously in 42 major emitter countries over 1998–2020 by extending the spatiotemporal production-theoretical decomposition models, and compared the different performances for forecasting CO2 emission by traditional time-series models and several machine learning models. Key findings as follows: first, drivers of CO2 emissions exhibit significant heterogeneity across countries where the effects of energy usage technology gap and CO2 emission technology gap were negative drivers for USA, South Korea, and the Czech Republic and potential energy intensity effect was the negative driver in countries like China, Russia, Japan, and India. Second, the effects of within-GDP per capita and within- population size were the important drivers affecting global CO2 emission difference. Third, general regression neural network achieved the best forecasting performance on average compared with other models in the study. The study highlights the importance of formulating climate policies based on heterogeneous technology and emission forecast modeling.

Twitter's capacity to forecast tourism demand: the case of way of Saint James

PurposeRadical changes in consumer habits induced by the coronavirus disease (COVID-19) pandemic suggest that the usual demand forecasting techniques based on historical series are questionable. This is particularly true for hospitality demand, which has been dramatically affected by the pandemic. Accordingly, we investigate the suitability of tourists’ activity on Twitter as a predictor of hospitality demand in the Way of Saint James – an important pilgrimage tourism destination.Design/methodology/approachThis study compares the predictive performance of the seasonal autoregressive integrated moving average (SARIMA) time-series model with that of the SARIMA with an exogenous variables (SARIMAX) model to forecast hotel tourism demand. For this, 110,456 tweets posted on Twitter between January 2018 and September 2022 are used as exogenous variables.FindingsThe results confirm that the predictions of traditional time-series models for tourist demand can be significantly improved by including tourist activity on Twitter. Twitter data could be an effective tool for improving the forecasting accuracy of tourism demand in real-time, which has relevant implications for tourism management. This study also provides a better understanding of tourists’ digital footprints in pilgrimage tourism.Originality/valueThis study contributes to the scarce literature on the digitalisation of pilgrimage tourism and forecasting hotel demand using a new methodological framework based on Twitter user-generated content. This can enable hospitality industry practitioners to convert social media data into relevant information for hospitality management.

A New Auto-Regressive Multi-Variable Modified Auto-Encoder for Multivariate Time-Series Prediction: A Case Study with Application to COVID-19 Pandemics.

The SARS-CoV-2 global pandemic prompted governments, institutions, and researchers to investigate its impact, developing strategies based on general indicators to make the most precise predictions possible. Approaches based on epidemiological models were used but the outcomes demonstrated forecasting with uncertainty due to insufficient or missing data. Besides the lack of data, machine-learning models including random forest, support vector regression, LSTM, Auto-encoders, and traditional time-series models such as Prophet and ARIMA were employed in the task, achieving remarkable results with limited effectiveness. Some of these methodologies have precision constraints in dealing with multi-variable inputs, which are important for problems like pandemics that require short and long-term forecasting. Given the under-supply in this scenario, we propose a novel approach for time-series prediction based on stacking auto-encoder structures using three variations of the same model for the training step and weight adjustment to evaluate its forecasting performance. We conducted comparison experiments with previously published data on COVID-19 cases, deaths, temperature, humidity, and air quality index (AQI) in São Paulo City, Brazil. Additionally, we used the percentage of COVID-19 cases from the top ten affected countries worldwide until May 4th, 2020. The results show 80.7% and 10.3% decrease in RMSE to entire and test data over the distribution of 50 trial-trained models, respectively, compared to the first experiment comparison. Also, model type#3 achieved 4th better overall ranking performance, overcoming the NBEATS, Prophet, and Glounts time-series models in the second experiment comparison. This model shows promising forecast capacity and versatility across different input dataset lengths, making it a prominent forecasting model for time-series tasks.

Enhancing Vegetable Price Forecasting Accuracy: A Hybrid SARIMA- DWT-GANN Approach

In the context of Malaysia's agricultural sector, the development of a robust vegetable price forecasting model holds paramount importance. The pricing dynamics of vegetables can significantly affect various stakeholders, including farmers, distributors, and consumers. The agricultural sector in Malaysia encounters persistent challenges such as supply-demand imbalances, seasonal variations, and market uncertainties, which can lead to income disparities for farmers and disruption in supply chains. Addressing these issues requires accurate and timely predictions of vegetable prices to enhance planning, resource allocation, and decision-making. This study introduces an innovative SARIMA-DWT-GANN hybrid model for vegetable price forecasting. By fusing the strengths of traditional time series modeling with the capabilities of neural networks, the proposed model offers a comprehensive solution to capture both linear and non-linear price patterns. The results demonstrate the superiority of the SARIMA-DWT-GANN model over the individual SARIMA model, as evident from correlation coefficients that closely approach unity and p-values confirming statistical significance. The model's ability to predict price changes has significant implications for making informed decisions throughout the agricultural supply chain. This research provides a robust forecasting tool that not only enhances market efficiency and profitability but also offers a promising solution to address the challenges in Malaysia's agricultural sector.

LQ45 Stock Portfolio Optimization by Considering Return Predictions Using the Support Vector Regression (SVR) Method

Investors require stock portfolio optimization to achieve desired outcomes. This process involves prediction to obtain optimal weights in the expected portfolio. Historically, portfolio optimization has primarily considered risk and expected value. However, a new approach has integrated return prediction into traditional time series models, enhancing the performance of portfolio optimization models. Machine learning has proven effective in predicting stock market trends, with many researchers applying these models in portfolio formation with satisfactory results. This study employs Support Vector Regression (SVR) method to integrate return prediction in portfolio formation, utilizing stock data from the LQ45 index. Test results indicate that return prediction using SVR method yields a Root Mean Square Error (RMSE) of 0.34973. Portfolios considering return prediction demonstrate superior performance compared to the LQ45 index, as evidenced by portfolio return averages, standard deviation, and Sharpe ratio.

Remaining lifespan prediction on multiple types of hard disks under conditions of data imbalance

Most of the previous regression models for predicting the remaining useful life (RUL) of hard disks primarily focused on failure data obtained through SMART monitoring. However, models that only considered failure data were not accurate enough. In order to address the issue of imbalanced sample quantities in different categories within the dataset, this paper proposes a method for predicting the RUL of hard disks under imbalanced conditions. It takes into account the imbalance issues in both SMART failure data and healthy data, as well as the imbalance in monitoring data attributes between hard disk drives and solid-state drives. The paper improves upon the traditional Transformer time series model and constructs a comprehensive and unified model for the RUL of hard disks across multiple categories. The proposed model achieves an RMSE score of 3.325, and has 66 % in MAE improved to the conventional models. The proposed model has promising prospects for future applications in predicting the remaining life of hard disks.

Modeling opening price spread of Shanghai Composite Index based on ARIMA-GRU/LSTM hybrid model.

In the dynamic landscape of financial markets, accurate forecasting of stock indices remains a pivotal yet challenging task, essential for investors and policymakers alike. This study is motivated by the need to enhance the precision of predicting the Shanghai Composite Index's opening price spread, a critical measure reflecting market volatility and investor sentiment. Traditional time series models like ARIMA have shown limitations in capturing the complex, nonlinear patterns inherent in stock price movements, prompting the exploration of advanced methodologies. The aim of this research is to bridge the gap in forecasting accuracy by developing a hybrid model that integrates the strengths of ARIMA with deep learning techniques, specifically Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) networks. This novel approach leverages the ARIMA model's proficiency in linear trend analysis and the deep learning models' capability in modeling nonlinear dependencies, aiming to provide a comprehensive tool for market prediction. Utilizing a comprehensive dataset covering the period from December 20, 1990, to June 2, 2023, the study develops and assesses the efficacy of ARIMA, LSTM, GRU, ARIMA-LSTM, and ARIMA-GRU models in forecasting the Shanghai Composite Index's opening price spread. The evaluation of these models is based on key statistical metrics, including Mean Squared Error (MSE) and Mean Absolute Error (MAE), to gauge their predictive accuracy. The findings indicate that the hybrid models, ARIMA-LSTM and ARIMA-GRU, perform better in forecasting the opening price spread of the Shanghai Composite Index than their standalone counterparts. This outcome suggests that combining traditional statistical methods with advanced deep learning algorithms can enhance stock market prediction. The research contributes to the field by providing evidence of the potential benefits of integrating different modeling approaches for financial forecasting, offering insights that could inform investment strategies and financial decision-making.

Empirical Evaluation of the Time Series Forecasting Method by Combining ARIMA with RBFNN under the Additive Model

Time series data is a series of values observed through repeated measurements at different times. Time series data is a type of data present in almost all different fields of life. Time series prediction is an significant problem in time series data mining. Accurate forecasting is crucial to support decision making in many areas of life. Therefore, improving the precision of time series predicting is a interesting mission for experts in this field. Many models for predicting time series have been proposed from traditional time series models as Auto Regressive Integrated Moving Average (ARIMA) model to artificial neural network (ANN) models. ARIMA is a linear model therefore it can only take the linear characteristics in time series. In contrast, Radial Basis Function Neural Network (RBFNN) is a non-linear model therefore it can not predict effectively seasonal or trend changes in time series. To combine the strengths of these two models, in this study, we experimentally evaluate the hybrid method between ARIMA and RBFNN on real time series data from different fields. Experimental results demonstrate that the combined method outperforms each model used individually in terms of accuracy.

MACHINE LEARNING FOR STOCK MARKET FORECASTING: A REVIEW OF MODELS AND ACCURACY

As financial markets become increasingly complex and dynamic, the application of machine learning (ML) techniques for stock market forecasting has garnered significant attention. This paper presents a comprehensive review of various ML models employed in the realm of stock market forecasting, focusing on their methodologies and the accuracy achieved in predicting market trends. The review begins by examining traditional time-series models such as autoregressive integrated moving average (ARIMA) and moving average convergence divergence (MACD) and their limitations in capturing the intricate patterns present in financial data. Subsequently, the discussion transitions to more advanced ML models, including support vector machines (SVM), artificial neural networks (ANN), and ensemble methods like random forests and gradient boosting. Each model's strengths and weaknesses are scrutinized in the context of stock market forecasting. The paper explores the pivotal role of feature selection and engineering in enhancing the predictive power of ML models. Feature sets encompassing financial indicators, macroeconomic variables, sentiment analysis from news articles, and social media data are analyzed for their impact on forecasting accuracy. Additionally, the incorporation of technical indicators and alternative data sources is explored as potential avenues to improve model robustness. A critical aspect of this review is the assessment of accuracy in predicting stock market movements. The evaluation is conducted through a comparative analysis of model performance metrics, including Mean Absolute Error (MAE), Mean Squared Error (MSE), and accuracy rates. The study also addresses the challenge of model overfitting and proposes strategies to mitigate this issue for more reliable predictions. This review provides a nuanced understanding of the landscape of ML models for stock market forecasting, highlighting the diverse approaches, challenges, and opportunities in the quest for improved accuracy. It contributes valuable insights for researchers, practitioners, and investors seeking to leverage the potential of ML in navigating the complexities of financial markets.
 Keywords: Machine Learning, Stock Market, Forecasting, Models, Review.