Moving Average Model Research Articles

Robust Estimation for Threshold Autoregressive Moving-Average Models

Threshold autoregressive moving-average (TARMA) models extend the popular TAR model and are among the few parametric time series specifications to include a moving average in a nonlinear setting. The state dependent reactions to shocks is particularly appealing in Economics and Finance. However, no theory is currently available when the data present heavy tails or anomalous observations. Here we provide the first theoretical framework for robust M-estimation for TARMA models and study its practical relevance. Under mild conditions, we show that the robust estimator for the threshold parameter is super-consistent, while the estimators for autoregressive and moving-average parameters are strongly consistent and asymptotically normal. The Monte Carlo study shows that the M-estimator is superior, in terms of both bias and variance, to the least squares estimator, which can be heavily affected by outliers. The findings suggest that robust M-estimation should be generally preferred to the least squares method. We apply our methodology to a set of commodity price time series; the robust TARMA fit presents smaller standard errors and superior forecasting accuracy. The results support the hypothesis of a two-regime non-linearity characterized by slow expansions and fast contractions.

PSATF-6mA: an integrated learning fusion feature-encoded DNA-6 mA methylcytosine modification site recognition model based on attentional mechanisms

DNA methylation is of crucial importance for biological genetic expression, such as biological cell differentiation and cellular tumours. The identification of DNA-6mA sites using traditional biological experimental methods requires more cumbersome steps and a large amount of time. The advent of neural network technology has facilitated the identification of 6 mA sites on cross-species DNA with enhanced efficacy. Nevertheless, the majority of contemporary neural network models for identifying 6 mA sites prioritize the design of the identification model, with comparatively limited research conducted on the statistically significant DNA sequence itself. Consequently, this paper will focus on the statistical strategy of DNA double-stranded features, utilising the multi-head self-attention mechanism in neural networks applied to DNA position probabilistic relationships. Furthermore, a new recognition model, PSATF-6 mA, will be constructed by continually adjusting the attentional tendency of feature fusion through an integrated learning framework. The experimental results, obtained through cross-validation with cross-species data, demonstrate that the PSATF-6 mA model outperforms the baseline model. The in-Matthews correlation coefficient (MCC) for the cross-species dataset of rice and m. musus genomes can reach a score of 0.982. The present model is expected to assist biologists in more accurately identifying 6 mA locus and in formulating new testable biological hypotheses.

NH4 Modelling with ARIMA and LSTM

AI-Mining is a prototype designed to detect various environmental gases, including CO2, NH3, NH4, and hydrogen, alongside temperature, pressure, and humidity. This study emphasizes the importance of modeling NH4 time series data due to its critical role in environmental and health monitoring. Accurate NH4 predictions facilitate early pollution detection and timely greenhouse gas control interventions. The study investigates the effectiveness of AI-Mining in detecting and predicting gas levels, focusing on data collection and analysis. Initial data analysis employed the Autoregressive Moving Average (ARIMA) model, specifically ARIMA (1,1,1), described by the equation yt = 0.0311 - 0.0750yt-1 + 0.3842εt-1. Despite its use, ARIMA's Root Mean Square Error (RMSE) performance was found lacking compared to more advanced methods. Given the classification of the obtained data as big data and time series, the Long Short-Term Memory (LSTM) method was also applied. The LSTM model initially used two layers with tanh and relu activation functions, and its performance was further explored by adding a third layer and varying the number of neurons (64, 128, and 256). The Adam optimizer was consistently used across all LSTM variations. Results indicated that increasing layers and neurons did not significantly impact LSTM's performance, with RMSE values around 0.023. However, LSTM consistently outperformed ARIMA in prediction accuracy, highlighting its robustness and reliability. Consequently, the study recommends using LSTM for predicting other recorded data in AI-Mining, underscoring its superiority in handling complex environmental datasets.

Conditional parametric bootstrap in GLARMA models

Due to situations where asymptotic results are not valid or when we do not know the true distributions of the data set, the bootstrap technique stands out as an effective alternative to make inferences about model parameters. In this context, we introduce the conditional parametric bootstrap, applied to time series data, specifically using the Generalized Linear Autoregressive Moving Average (GLARMA) model. Our interest is to use the procedure to correct possible biases and construct confidence intervals and hypothesis testing for the parameters of the GLARMA model. We conducted simulation studies to evaluate the performance of the procedure in finite samples. The results show that the proposed technique is promising and can be applied in practical situations. An example is the analysis of the monthly number of chronic obstructive pulmonary disease (COPD) cases, which illustrates the potential of the proposed approach in a real situation.

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.

Characteristics of Machine Learning-based Univariate Time Series Imputation Method

Handling missing values in univariate time series analysis poses a challenge, potentially leading to inaccurate conclusions, especially with frequently occurring consecutive missing values. Machine Learning-based Univariate Time Series Imputation (MLBUI) methods, utilizing Random Forest Regression (RFR) and Support Vector Regression (SVR), aim to address this challenge. Considering factors such as time series patterns, missing data patterns, and volume, this study explores the performance of MLBUI in simulated Autoregressive Integrated Moving Average (ARIMA) datasets. Various missing data scenarios (6%, 10%, and 14%) and model scenarios (Autoregressive (AR) models: AR(1) and AR(2); Moving Average (MA) models: MA(1) and MA(2); Autoregressive Moving Average (ARMA) models: ARMA(1,1) and ARMA(2,2); and Autoregressive Integrated Moving Average (ARIMA) models: ARIMA(1,1,1) and ARIMA(1,2,1)) with different standard deviations (0.5, 1, and 2) were examined. Five comparative methods were also used in this research, including Kalman StructTS, Kalman Auto-ARIMA, Spline Interpolation, Stine Interpolation, and Moving Average. The research findings indicate that MLBUI performs exceptionally well in imputing successive missing values. The results of this study indicate that the performance of MLBUI in imputing consecutive missing values, based on MAPE, yielded values of less than 10% across all scenarios used.

Modeling Survey Time Series Data with Flow-Observed CARMA Processes

Published survey data often are delivered as estimates computed over an epoch of time. Customers may desire to obtain survey estimates corresponding to epochs, or time points, that differ from the published estimates. This “change of support” problem can be addressed through the use of continuous-time models of the underlying population process, while taking into account the sampling error that survey data is subject to. The application of a Continuous AutoRegressive Moving Average (CARMA) model is investigated as a tool to provide change of support applications, thereby allowing interpolation for published survey estimates. A simulation study provides comparisons of competing estimation methods, and a synthetically constructed data set is developed in order to elucidate real data applications. The proposed method can be successful for change of support problems, despite modeling challenges with the CARMA framework.

Modelling Motor Insurance Claims Using Arma Model And Appropriate Statistical Distribution Of Insurance Companies, Machakos County, Kenya

The prediction of future claim amounts based on past or present claims is a challenging task for insurance companies. According to research carried out by Michael Barth and David L. Eccles (2009); claim costs constitute large proportion of insurers expenses in evaluating the premium growth on the loss ratio of property causality insurers. Various statistical techniques have been employed to analyze claims data and assess its impact on productivity. Autoregressive (AR), Moving Average (MA), and Auto Regressive Moving Average (ARMA) processes are commonly used to model the temporal dependencies and trends in the data. These techniques enable insurance companies to understand the dynamics of claim incidence and their effects on productivity over time (Philip J, 2016). Advanced modeling techniques such as generalized additive models, negative binomial regression, and multinomial logit models have been used to evaluate claims processes, predict claim occurrence, and model claim severity (Renshaw A.E, 1994; Edward. W Frees et al., 2008; Daiane Aparecida Zanetti et al., 2006). These models consider various factors such as driver age, gender, vehicle type, and no claim discount to assess their impact on claim incidence and productivity. Previous studies have overlooked the comprehensive examination of the impact of insurance claims on productivity on insurance companies. Therefore, this research aims to address this gap by investigating the factors associated with insurance motor claims and their effects on the productivity of insurance companies by employing statistical techniques like appropriate probability distribution, incorporation of moving average technique, Autoregressive process and ARMA process, as identified by industry professionals and experts. The results obtained from the real-life data for distinctive 11 years indicated the potential strength of ARMA models to provide insurance company's claim amount prediction that could assist the companies to have better planning. The findings of this study, if adopted by various keynote players in the insurance and regulatory sectors, can be highly beneficial in formulating appropriate policies to enhance the management of motor insurance claims and improve the productivity of insurance companies.

News Popularity Prediction in West Sumatera Using Autoregressive Integrated Moving Average

The increasing public interest in reading online news is undoubtedly a challenge for news portals as online news providers. Therefore, this research was conducted to predict news popularity in West Sumatra through the FajarSumbar.com news portal using the Autoregressive Integrated Moving Average (ARIMA) model. This research aims to develop a forecasting model that can assist in estimating the popularity of each news category so that news portals can devise more effective content strategies. The data used in this study includes the number of monthly news impressions from March 2021 to June 2024, which are grouped into various categories such as Religion Culture, Industrial Economics, Criminal Law, etc. Using the ARIMA method, which can handle time series data and overcome data non-stationarity problems through differencing and the use of grid search in optimization to find the best parameters based on the lowest evaluation metric. The results show that the ARIMA model can provide reasonably accurate predictions, although the level of accuracy varies between categories. The Mean Absolute Percentage Error (MAPE) values obtained are as follows: Religion Culture 26%, Industrial Economy 29%, Criminal Law 29%, Health 40%, Sports 38%, Tourism Entertainment 26%, Education 27%, Government Politics 31%, Social Environment 27%, and Technology 51%. The Technology and Health news categories show higher error rates than others, while Religion Culture and Tourism Entertainment have better accuracy rates. Thus, the ARIMA model can be used to predict future trends in news popularity, helping editors plan content strategies that are more relevant and interesting to readers. However, improvements are needed for news categories that have high variability.

Multi-Scenario land cover changes and carbon emissions prediction for peak carbon emissions in the Yellow River Basin, China

Research on future land cover changes and carbon emissions is essential for effective land resource management and developing feasible carbon mitigation strategies. This study focused on the Yellow River Basin and employed the Future Land Use Simulation (FLUS) and Autoregressive Integrated Moving Average (ARIMA) models to project future land cover and carbon emissions. Additionally, bivariate spatial autocorrelation was utilized to analyze the relationship between them. Key findings are as follows: 1) Historically, the Yellow River Basin has experienced an expansion in construction land, forests, grasslands, and water, while cropland and unused land have diminished. Notably, construction land displayed the most significant changes, whereas grasslands showed minimal modification. Looking ahead, both the ecological protection and inertial development scenarios exhibit consistent trends with historical patterns across the land type categories. In contrast, the economic priority development scenario forecasts an increase in construction land, cropland, and grasslands, indicating a distinct shift compared to the other scenarios. However, the ecological protection scenario proves to be more sustainable. 2) In the absence of intervention, the simulated carbon emissions from construction land throughout the basin display a linear increase across various scenarios, with provincial-level variations showing an increase from southwest to northeast. However, Henan and Sichuan are expected to experience slower reductions in carbon emissions, compared to other projections. There is a notable positive correlation between carbon emissions and the comprehensive index, indicating that regions with high emissions typically experience substantial land and economic development. 3) Energy consumption projections for 2030 and 2060 indicate that to align with China’s carbon goals, it is essential to reduce energy consumption and adjust the fossil to non-fossil fuel ratio to reduce carbon emissions. Substituting coal with clean energy and enhancing energy efficiency will be more effective for achieving low-carbon emission targets. In summary, this study provides significant guidance for China’s ecological conservation, low-carbon emission strategies, and global carbon emission control efforts.

Forecasting Daily Radiotherapy Patient Volumes in a Tertiary Hospital Using Autoregressive Integrated Moving Average (ARIMA) Models.

The purpose is to predict the volume of patients treated daily with radiotherapy using the autoregressive integrated moving average (ARIMA) model. In this retrospective study, data from the billing records detailing daily radiotherapy treatment sessions were extracted from the Hospital Information System and analyzed. The study included all patients treated from January 2004 to December 2022. The analysis was divided into two parts: First, the data were summarized using descriptive statistics. Second, time series forecasting with the implementation of an ARIMA model for estimating patient volumes. For the ARIMA modeling process, the Akaike Information Criterion (AIC) was used for classical model optimization. The Mean Absolute Percentage Error (MAPE) was used for evaluating between different models. Residual analysis was performed in each model using the Ljung-Box test, Jarque-Bera test, and heteroskedasticity test to identify autocorrelation, normal distribution, and variances that could undermine the reliability of the model. A total of 895,808 radiotherapy sessions were included in the study. The median number of radiotherapy sessions per day was 181 (150, 205). A clear transition to more modern radiotherapy equipment, particularly the Truebeam accelerator, was observed, indicating a growing dependency on advanced techniques such as volumetric-modulated arc therapy (VMAT), stereotactic body radiation therapy (SBRT), and stereotactic radiosurgery (SRS). The best ARIMA model predicted an increase in demand, projecting an average daily patient volume of 279.40 by 2030. The study highlights the need for advanced forecasting methodologies in healthcare resource planning and emphasizes the importance of considering environmental and external factors for effective and accurate resource allocation strategies.

Forecasting stock index closing points using ARIMA-GARCH with a rolling data window

This research develops and examines the efficacy of a hybrid ARIMA-GARCH model, augmented by a rolling data window approach, to enhance the accuracy of stock index prediction, specifically focusing on the NEPSE index. Accurate predictions of stock indices are of paramount importance to investors, analysts and policy makers to navigate and circumvent the market uncertainties. The AutoRegressive Integrated Moving Average (ARIMA) model captures linear trends and temporal dependencies in time series data, while the General AutoRegressive Conditional Heteroskedasticity (GARCH) model addresses volatility clustering—ubiquitous character of financial time series—thereby providing a comprehensive framework for prediction. Utilizing a dataset comprised of daily closing points of NEPSE index approximately three years and nine months, the study identifies the ARIMA (5,1,0)-GARCH (1,1) model as optimal fit, upon integrating 180-day rolling data window. This model achieved a Mean Percentage Error of -0.0058% and a correlation of 0.995, which is indicative of superior fit to the underlying time series data. These findings underscore the hybrid model’s capacity to adaptively respond to dynamic market conditions and acclimatize prediction parallel to most recent market trends and volatility. This research is useful for optimizing investment strategies for those invested in Nepalese stocks. Also, this research lays a foundational framework for future investigations into application of this advanced forecasting method in other emerging markets, financial instruments and indices.

Assessing the Predictive Capabilities of Autoregressive Integrated Moving Average and Linear Regression Models for Acute Changes in Clinical and Selected Laboratory Parameters in Children After Cardiac Surgery in the ICU

(1) Background: The main objective of this research was to assess the clinical factors related to the condition of pediatric patients with congenital heart defects after they underwent intensive care unit surgery. The information was gathered from the Congenital Heart Disease Surgery Unit at the National Heart Foundation Hospital and Research Institute in Dhaka, Bangladesh. We gathered and examined data from 288 ICU patients. Patients under the age of twelve who required more than a 24-h ICU stay were selected. (2) Methods: The dependent and independent variables were chosen in advance based on expert opinion. The relationships between these pre-specified ICU parameters were determined using the Pearson correlation model and assessed through linear regression and ARIMA modeling to predict subsequent acute changes in the patients’ ICU statuses. (3) Results: A statistically significant relationship (p value < 0.001) was found between CVP and BP (95% CI = 0.2113; 0.353 r = 0.2841249) and between PEEP and FiO2 (95% CI = 0.6992; 0.770 r = 0.7367744). Although the relationships between pH and PO2 were minor (95% CI = 0.161; 0.308 r = 0.2362575), they were statistically significant. The parameters considered statistically significant (p < 0.001) were chosen for forecasting. In this work, the linear regression model and the ARIMA model used the parameters BP, FiO2, and PO2 for prediction. We forecasted the patients’ statuses for the next hour. It was found that the ARIMA model had a lower error rate than the linear regression model. (4) Conclusions: This study helps identify the important parameters for predicting and monitoring patients’ statuses in the ICU, with the ultimate goal of providing physicians with an early warning system to anticipate deterioration in clinical and biochemical parameters. The ability to accurately forecast future patients’ conditions can enable proactive, targeted interventions, potentially improving outcomes and reducing the risk of adverse events.

Modelling net primary productivity using near real-time land cover data and soil moisture information

ABSTRACT The objective of this study is to develop a methodological approach to assess ecosystem functioning in terms of carbon uptake and sequestration. To achieve this, a model was developed to forecast Net Primary Productivity (NPP) and tested in a rural area, the Thrace region (NE Greece). Monthly net photosynthesis data from July 2015 to February 2024 was estimated using the MODerate Resolution Imaging Spectroradiometer (MODIS) Aqua MYD17A2H v061 dataset for the study area. This dataset was used as the dependent variable in an Autoregressive Integrated Moving Average (ARIMA) model. The predictive ability of monthly Land Cover (LC) and Soil Moisture (SM) data at various time lags was examined, and variables found to be significant predictors were introduced into the model as external predictors. The study found that crop areas and SM conditions exert the greatest influence on NPP, with areas containing trees exerting a lesser impact. Additionally, NPP was estimated for seven different LC and SM scenarios from March 2023 to February 2024.

Forecasting Orange Juice Futures: LSTM, ConvLSTM, and Traditional Models Across Trading Horizons

This study evaluated the forecasting accuracy of various models over 5-day and 10-day trading horizons to predict the prices of orange juice futures (OJ = F). The analysis included traditional models like Autoregressive Integrated Moving Average (ARIMA) and advanced neural network models such as Long Short-Term Memory (LSTM), Recurrent Neural Network (RNN), Backpropagation Neural Network (BPNN), Support Vector Regression (SVR), and Convolutional Long Short-Term Memory (ConvLSTM), incorporating factors like the Commodities Index and the S&P500 Index. We employed loss function metrics and various tests to assess model performance. The results indicated that for the 5-day horizon, the LSTM and ConvLSTM consistently outperformed the other models. LSTM achieved the lowest error rates and demonstrated superior capability in capturing temporal dependencies, especially in single-factor and S&P500 Index predictions. ConvLSTM also performed strongly, effectively modeling spatial and temporal data patterns. In the 10-day horizon, similar trends were observed. LSTM and ConvLSTM models had significantly lower errors and better alignment with actual values. The BPNN model performed well when all factors were included, and the SVR model maintained consistent accuracy, particularly for single-factor predictions. The Diebold–Mariano (DM) test indicated significant differences in forecasting accuracy, favoring advanced neural network models. In addition, incorporating multiple influencing factors further improved predictive performance, enhancing investment outcomes and reducing risk.

Electric bus fleet transition: assessment approach considering economic and environmental impacts, and its application

In our society, global warming is considered one of the most serious problems. According to scientists, the world has been warmed by 3 degrees per year, which will be catastrophic to our world. To reduce CO2 emission, an electric bus is one way to solve the problem. In this article, we use four different models: Multiple Linear Regression (MLR), Autoregressive Integrated Moving Average Model (ARIMA), Ecological Assessment Model, Bus Fleet Replacement Financial Model, and Integer Programming Model to determine the number of carbon emissions, the least money that government need to spend on transitions, and future blueprint; these help to predict the overall benefits for countries turn into absolutely electric bus society. Our research stands from the sustainable point of view; we view better environment as the goal. By applying these models to three different countries: London, and Toronto, and Philadelphia which is our main focus, we find out that the air quality will be increased by reducing different kinds of pollution. Moreover, by constructing a ten-year blueprint, we find out the best way to spend least money and make the environment gradually become better.

The impact of new VAT enforcement on financial performance: Evidence from Saudi Arabia non-financial listed companies using the event study and ARMA model

This paper examines the influence of the new 15 % Value Added Tax (VAT) enforcement on non-financial listed companies in Saudi Arabia. By comparing financial data from 2019, before the VAT implementation and the COVID-19 pandemic, with data from 2020, during both the VAT increase and the pandemic, the research aims to uncover the consequences of this tax policy change. Utilizing charts, tables, and an event study analysis approach with the Autoregressive Moving Average (ARMA) model, we investigated key financial indicators such as Shareholders' Equity (SE), Total Income (TI), Total Revenues (TR), Net Income (NI), Total Expenses (TE), Other Changes in Operating Activity (COA), and Cash at the End of the Period (CEP). The findings reveal significant financial impacts, highlighting companies' challenges due to the increased tax burden. Shareholders' Equity dropped from an average of 11.55 million to 4.57 million, and Total Income and Total Revenues decreased from approximately 1.73 million to 316,234 and from 1.95 million to 331,605, respectively. Net Income sharply declined from 406,109 to 74,624, and Total Expenses decreased from about 1.18 million to 223,495. Other Changes in Operating Activity shifted from a negative mean of −168,936 to a positive 523,520, and Cash at the End of the Period fell from 1.99 million to 666,663. These results suggest that the VAT increase has significantly strained companies' financial performance, emphasizing the need to consider such tax reforms carefully, particularly in developing countries. The study concludes that while the VAT increase aims to enhance government revenues, it imposes substantial financial strain on companies, potentially leading to long-term economic repercussions. Policymakers should consider these implications and implement practical measures to optimize the VAT system and support business resilience in the face of such reforms.

Calibration of Arctic ice core bromine enrichment records for past sea ice reconstructions

Bromine in ice cores has been proposed as a qualitative sea ice proxy to produce sea ice reconstructions for the polar regions. Here we report the first statistical validation of this proxy with satellite sea ice observations by combining bromine enrichment (with respect to seawater, Brenr) records from three Greenlandic ice cores (SIGMA-A, NU and RECAP) with satellite sea ice imagery, over three decades. We find that during the 1984–2016 satellite-era, ice core Brenr values are significantly correlated with first-year sea ice formed in the Baffin Bay and Labrador Sea supporting that the gas-phase bromine enrichment processes, preferentially occurring over the sea ice surface, are the main driver for the Brenr signal in ice cores. Moreover, in assessing Brenr's capability to record historical sea ice variability, we compare 20th-century Arctic Sea ice historical and proxy records with our reconstructions, based on an autoregressive–moving-average (ARMA) model, finding overall good agreement. While further enhancements are warranted, including site-specific calibrations and a comprehensive investigation into bromine transport-related concerns, this study presents a new method to quantitatively reconstruct past seasonal sea ice variability through bromine enrichment in ice cores.

Fifteen Years of Non-Hodgkin Lymphoma in an Indonesian National Referral Hospital: Epidemiologic Trends and Diagnostic Challenges.

The global burden of lymphoma is substantial because of the increase in its incidence in recent decades. However, disease characteristics vary across different geographical locations. Numerous immunohistochemistry markers and molecular studies are essential to determine lymphoma diagnosis and prognosis. This poses significant challenges in developing countries with limited health care resources. This large-scale study assesses the frequency of non-Hodgkin lymphoma (NHL) in Indonesia over the past 15 years, analyses its clinicopathologic features, and predicts future trends. This retrospective study collected lymphoma patients diagnosed at the Department of Anatomical Pathology Dr. Cipto Mangunkusumo National Central General Hospital, Indonesia, from 2009 until 2023. All lymphoma diagnoses were confirmed by using ancillary tools classified as an enhanced lymphoma panel according to a resource-stratified guideline. We analyzed the clinicopathologic features of each NHL type and further applied the Autoregressive Integrated Moving Average model to predict future incidence trends. The study consisted of 7,368 NHL patients. Among these, B-cell lymphomas accounted for 90.6%, with diffuse large B-cell lymphoma being the most prevalent subtype (68.8%), followed by follicular lymphoma (8.8%) and marginal zone lymphoma (5.8%). Extranodal natural killer/T-cell lymphoma, nasal type, is the most common T-cell lymphoma found (26.3%). All types of lymphoma were found to be more common in males (57.7%). Extranodal involvement, particularly in the tonsil and upper respiratory tract, was frequently observed. Projection analysis indicates a steady increase in lymphoma patients in the future. This study highlights the distribution and burden of NHL in Indonesia over 15 years. The overall epidemiologic pattern of NHL in this study aligns with the results observed in other Asian countries. The rising incidence of lymphoma requires improved health care infrastructure and prevention strategies.

Hybrid Long Short-Term Memory Wavelet Transform Models for Short-Term Electricity Load Forecasting

To ensure the constant availability of electrical energy, power companies must consistently maintain a balance between supply and demand. However, electrical load is influenced by a variety of factors, necessitating the development of robust forecasting models. This study seeks to enhance electricity load forecasting by proposing a hybrid model that combines Sorted Coefficient Wavelet Decomposition with Long Short-Term Memory (LSTM) networks. This approach offers significant advantages in reducing algorithmic complexity and effectively processing patterns within the same class of data. Various models, including Stacked LSTM, Bidirectional Long Short-Term Memory (BiLSTM), Convolutional Neural Network—Long Short-Term Memory (CNN-LSTM), and Convolutional Long Short-Term Memory (ConvLSTM), were compared and optimized using grid search with cross-validation on consumption data from Lome, a city in Togo. The results indicate that the ConvLSTM model outperforms its counterparts based on Mean Absolute Percentage Error (MAPE), Root Mean Squared Error (RMSE), and correlation coefficient (R2) metrics. The ConvLSTM model was further refined using wavelet decomposition with coefficient sorting, resulting in the WT+ConvLSTM model. This proposed approach significantly narrows the gap between actual and predicted loads, reducing discrepancies from 10–50 MW to 0.5–3 MW. In comparison, the WT+ConvLSTM model surpasses Autoregressive Integrated Moving Average (ARIMA) models and Multilayer Perceptron (MLP) type artificial neural networks, achieving a MAPE of 0.485%, an RMSE of 0.61 MW, and an R2 of 0.99. This approach demonstrates substantial robustness in electricity load forecasting, aiding stakeholders in the energy sector to make more informed decisions.