Outliers In Data Research Articles

P-267. Methodological Framework for Estimating Infections Prevented by Infection Control Services

Abstract Background This report aims to propose a methodology for estimating the number of infections prevented by an Infection Control Service, presenting observed outcome estimates at a University Hospital in 2023.Table 1Expected Hospital Infections for 2023, Observed Infections, and Infections Prevented in 2023, Month-by-Month, for the ICU on the 3rd FloorExpected Hospital Infections for 2023, Observed Infections, and Infections Prevented in 2023, Month-by-Month, for the ICU on the 3rd Floor Methods Prevented infection is estimated by the Infectometer, a real-time system that provides information on the expected weekly and monthly incidence of specific infections. To estimate the number of infections prevented each year, the expected monthly HAI count is calculated based on endemic curves from the immediately preceding year. This figure is then compared with the actual number of infections observed month-to-month, yielding an estimate of the number of infections avoided. To calculate Infectometer parameters: (1) construct the Statistical Process Control (SPC) for each infection or microorganism; (2) estimate the average month (m) and standard deviation (s) of the infection cases, excluding outlier data, i.e., data from months with rates higher than the maximum control limit; (3) calculate the Monthly Expected Incidence (MEI), assuming that the incidence of new infections (NI) follows a normal distribution. The 50th percentile from the normal distribution is used when the objective is to estimate the number of prevented HAIs.Table 2Expected Hospital Infections for 2023, Observed Infections, and Infections Prevented in 2023, Month-by-Month, for the ICU on the 1st FloorExpected Hospital Infections for 2023, Observed Infections, and Infections Prevented in 2023, Month-by-Month, for the ICU on the 1st Floor Results The number of expected infections per month in 2023, by site, was estimated based on the Infectometer and parameters from the previous year (2022), with 13 surgical site infections, 12 tracheobronchitis, 7 urinary tract infections, 6 primary bloodstream infections, 6 pneumonia, and 4 other infections. Based on the number of infections observed in 2023, there were 43 infections on the 2nd Floor, 128 on the 4th Floor, 93 in the ICU on the 1st Floor, and 42 in the ICU on the 3rd Floor, totaling 306 infections prevented.Table 3Expected Hospital Infections for 2023, Observed Infections, and Infections Prevented in 2023, Month-by-Month, for the Inpatient Units on the 4th FloorExpected Hospital Infections for 2023, Observed Infections, and Infections Prevented in 2023, Month-by-Month, for the Inpatient Units on the 4th Floor Conclusion A methodology was presented for estimating the number of infections prevented by the actions of a multimodal, multifaceted, and multiprofessional program managed by the Infection Control Service (ICS), whose main objective is to control and prevent healthcare-associated infections (HAIs) at a university hospital. Based on this methodology, 306 infection cases were prevented in 2023, saving at least $125,154 over one year. Additionally, 70 deaths were prevented in 2023.Table 4Expected Hospital Infections for 2023, Observed Infections, and Infections Prevented in 2023, Month-by-Month, for the Inpatient Units on the 2nd FloorExpected Hospital Infections for 2023, Observed Infections, and Infections Prevented in 2023, Month-by-Month, for the Inpatient Units on the 2nd Floor Disclosures All Authors: No reported disclosures

Clustering of Districts Based on Infrastructure Indicators Using K-Means And Average Linkage Methods

Aceh Jaya is one of the districts in Aceh Province that has great potential in various sectors, such as agriculture, plantations, and tourism. In the Aceh Jaya District Development Plan (RPK) Book 2023-2026, it is stated that the unequal distribution of infrastructure development in Aceh Jaya District in 2023-2026 has resulted in low investment competitiveness and decreased economic performance. Because of these conditions, this study clusters regions in Aceh Jaya District based on infrastructure indicators so that groups of sub-districts are obtained based on the level of infrastructure development. The method used in this study is clustering analysis by comparing it with two clustering methods, namely K-means and average linkage, which are validated based on the silhouette coefficient value to see which cluster is the best. The results obtained are the Kaiser-Mayer-Olkin (KMO) value of 0.611, which indicates that the sample is representative of a total of eight variables; for the zscore value, there is no outlier data, which means that the data of the eight infrastructure variables can be used entirely. Meanwhile, the clustering results using the K-Means method resulted in 3 clusters, and with the Average Lingkage method resulted in 2 clusters. This means that Aceh Jaya District can be clustered into 2 clusters, as seen from the silhouette coefficient value of 0.736. The conclusion that can be drawn from this research is that infrastructure development has not been evenly distributed in the 9 existing sub-districts, so that strategies and priorities are needed in infrastructure development in Aceh Jaya Regency.

Age predicts peak gamma frequency and N1 amplitude of visual evoked potential

The current study investigated whether the age of healthy adults could predict the peak gamma frequency and the peak amplitudes of VEP components (N1, P2). 49 healthy participants (aged between 19 and 52 years) underwent EEG recordings during a visual task eliciting clear gamma frequency oscillations and VEP activities. After eliminating noisy and outlier data, data from 41 participants were analysed using simple linear regression. The results indicated that age was a significant predictor of peak gamma frequency and the peak amplitude of VEP-N1 but not the peak amplitude of VEP-P2. Age was negatively associated with peak gamma frequency and the peak amplitude of VEP-N1. These findings support previous research indicating that ageing is associated with decreased cortical inhibition, highlighting the importance of GABA in maintaining cortical E-I balance.

Early Detection of Chronic Kidney Disease Using a Machine learning Approach for Better Results

Chronic Kidney Disease has also become one of the serious public health threats worldwide because of its chronic nature and potential progression to End-Stage Renal Disease if left untreated or undiagnosed. This paper aims to utilize machine learning techniques for the early detection of CKD using algorithms such as KNeighborsClassifier, Gaussian Naive Bayes, Support Vector Machine, and Random Forest. Models applied to a data set that contains clinical attributes such as age, blood pressure, and albumin levels after preprocessing for data quality normalization, feature selection, and outlier management. The performance metrics of the developed models were based on the accuracy, precision, recall, and F1 score; Random Forest has the highest diagnostic accuracy at 98.06%. The results show that machine learning models, in this case, Random Forest, are useful for prediction of CKD and may support non-invasive, low-cost early detection in clinical practice. Future work will focus on ensemble and deep learning approaches in order to enhance the robustness of models and their applicability in healthcare. Index Terms—Chronic kidney disease , Machine Learning ,KNeighborsClassifier, Support vector machine, Health Infor- matics, Supervised Learning, Medical Data Preprocessing, Early Disease Detection.

Collaborative Integrated Navigation for Unmanned Aerial Vehicle Swarms Under Multiple Uncertainties

UAV swarms possess unique advantages in performing various tasks and have been successfully applied across multiple scenarios. Accurate navigation serves as the foundation and prerequisite for executing these tasks. Unlike single UAV localization, swarms enable the sharing and propagation of precise positioning information, which enhances overall swarm localization accuracy but also introduces the issue of uncertainty propagation. To address this challenge, this paper proposes an integrated navigation and positioning method that models, propagates, and mitigates uncertainties. To tackle the issue of uncertainty in information quality caused by outliers in external correction data, a robust integrated navigation method for nonlinear systems is derived based on a normal gamma distribution model. Considering uncertainty propagation, a statistical linearization model for nonlinear systems is developed. Building upon this model, an augmented measurement nonlinear least squares positioning method is applied, achieving further improvements in localization accuracy. Simulation experiments demonstrate that the proposed method, which thoroughly accounts for the effects of multiple uncertainties, can achieve robust tracking and provide relatively accurate positioning results.

MutaSwarmClus: enhancing data clustering efficiency with mutation-enhanced swarm algorithm

Data clustering is a fundamental technique in data mining, pivotal for various applications such as statistical analysis and data compression. Traditional clustering algorithms often struggle with noisy or high-dimensional datasets, hindering their efficacy in addressing real-world challenges. In response, this research introduces MutaSwarmClus, a novel hybrid metaheuristic algorithm that combines Mouth Brooding Fish (MBF), Ant Colony Optimization (ACO), and mutation operators to enhance clustering quality. MutaSwarmClus intends to adaptively control the exploration and exploitation phases of the solution space, solve issues with local optima and changes in the distribution of available data. Moreover, it incorporates an Iterated Local Search (ILS) to refine solutions and avoiding getting stuck in local optima. MutaSwarmClus therefore increases the robustness of the clustering process by incorporating controlled randomness through mutation operators in order to handle noisy and outlier data points well. According to the contributions analysis, the proposed algorithm improves the clustering solution with the combined system of MBF, ACO, and mutation operators, which enables the mechanism of exploration and exploitation in the process of information search. As shown through the results of experimental studies, MutaSwarmClus has high performance when used with various benchmarks, and outperforms or performs as well as or better than compared to other clustering algorithms such as K-means, ALO, Hybrid ALO, and MBF. It achieves an average error rate of only 10%, underscoring its accuracy in clustering tasks. The utilization of MutaSwarmClus offers a solution to the existing problems in clustering large datasets in terms of scalability, efficiency and accuracy. Possible directions for future work can continue to optimize the model parameters of the algorithm and study its adaptability in dynamic conditions and with large amounts of data.

Out-of-Distribution Detection via outlier exposure in federated learning.

Among various out-of-distribution (OOD) detection methods in neural networks, outlier exposure (OE) using auxiliary data has shown to achieve practical performance. However, existing OE methods are typically assumed to run in a centralized manner, and thus are not feasible for a standard federated learning (FL) setting where each client has low computing power and cannot collect a variety of auxiliary samples. To address this issue, we propose a practical yet realistic OE scenario in FL where only the central server has a large amount of outlier data and a relatively small amount of in-distribution (ID) data is given to each client. For this scenario, we introduce an effective OE-based OOD detection method, called internal separation & backstage collaboration, which makes the best use of many auxiliary outlier samples without sacrificing the ultimate goal of FL, that is, privacy preservation as well as collaborative training performance. The most challenging part is how to make the same effect in our scenario as in joint centralized training with outliers and ID samples. Our main strategy (internal separation) is to jointly train the feature vectors of an internal layer with outliers in the back layers of the global model, while ensuring privacy preservation. We also suggest an collaborative approach (backstage collaboration) where multiple back layers are trained together to detect OOD samples. Our extensive experiments demonstrate that our method shows remarkable detection performance, compared to baseline approaches in the proposed OE scenario.

The Role of Company Size Moderating the Effect of Capital Structure and Profitability to Firm Value in Consumer Non-Cyclicals Businesses on the Indonesian Stock Exchange

Aims: Study this is about the firm value, aiming to test company size to moderate the influence of capital structure and profitability on the firm value in consumer non-cyclical business listed on the Indonesia Stock Exchange (IDX). Study Design: The design of this research study is correlational. Place and Duration of Study: Consumer non-cyclical business on the Indonesia Stock Exchange (IDX) issuers in 2018-2022. Methodology: This study's population was 122 non-cyclical consumer companies listed on the Indonesia Stock Exchange from 2018 to 2022. From the population, a sample of 56 was selected based on the purposive sampling method with 5 years of observation (2018-2022), so the number of samples was 280. However, there were 18 outlier data, so the final sample was 262. Data were collected using documentary methods and analyzed using a quasi-moderation model, which was processed with SPSS 24. Results: The study's results indicate that company size can moderate the effect of capital structure and profitability on company value. This means that capital structure and profitability on a larger company scale can strengthen the increase in company value. However, partially or individually, company size, capital structure, and profitability do not affect company value. Conclusion: The top-based results of the contribution study support the signalling theory and prove that empirical improvement in the size of a company can moderately influence the capital structure and profitability on increased firm value.

Third trimester fetal 4D flow MRI with motion correction.

To correct maternal breathing and fetal bulk motion during fetal 4D flow MRI. A Doppler-ultrasound fetal cardiac-gated free-running 4D flow acquisition was corrected post hoc for maternal respiratory and fetal bulk motion in separate automated steps, with optional manual intervention to assess and limit fetal motion artifacts. Compressed-sensing reconstruction with a data outlier rejection algorithm was adapted from previous work. Pre- and post-motion correction comparison included qualitative visibility of vasculature on phase-contrast MR angiograms (five-point Likert scale), conservation of mass of the aortic isthmus, ductus arteriosus, and descending aorta, and coefficient of variation of flow along the descending aorta. Twenty-nine third trimester acquisitions were performed for 15 healthy fetuses and two patients with postnatally confirmed aortic coarctation during a single examination for each participant. Only 15/27 (56%) of all volunteers and 1/2 (50%) of all patient precorrection acquisitions were suitable for flow analysis. Motion correction recovered eight "failed" acquisitions, including one patient, with 24/29 (83%) suitable for flow analysis. In the 15 viable uncorrected volunteer acquisitions, motion correction improved phase-contrast MR angiograms visibility significantly in the ductus arteriosus (from 4.0 to 4.3, p = 0.04) and aortic arch (3.7 to 4.0, p = 0.03). Motion correction improved conservation of mass to a median (interquartile range) percent difference of 5% (9%) from 14% (24%) with improvement shown in 14/15 acquisitions (p = 0.002), whereas coefficient of variation changes were not significantly different (uncorrected: 0.15 (0.09), corrected: 0.11 (0.09), p = 0.3). Motion correction compensated for maternal and fetal motion in fetal 4D flow MRI data, improving image quality and conservation of mass.

WHO IS “OLD” IN AN AFFLUENT RETIREMENT COMMUNITY: QUALIFYING STIGMA THROUGH OUTLIER DATA

Abstract The purpose of this study has been to compare and potentially reconcile professional constructs defining who is an older adult with lived experience of aging into 80+ decades. In this anthropological field research project, the study site is an affluent, Midwestern retirement community in which the average age is 85. Data were primarily collected through participant observation and semi-structured interviews during several data collection periods from 1 week to 2 months between July 2021 and December 2023. Despite optimal access to resources and identities of privilege, findings show how residents confronted “deep old age” as a highly stigmatized experience and social identity. Demographic and other quantitative data were collected from administrative data. These data helped qualify normative responses from research participants, such as a perceived “rash of deaths,” during the Covid-19 pandemic. In addition, qualitative data provided richness of detail and contextualization that numbers could not. In the objective concerns of a facility caring for older adults and the subjective experiences of residents confronting fears of aging, outlier data included examples of coping and social interactions that did not fit professional nor peer expectations of aging as stigmatizing or debilitating. Such examples inspired hope of alternative constructs that transformed ageist fears of age-related change. Residents who inspired included those who were unusually able to accept frailty, falls, and disability without resignation. In dialogue with research participants, “old” age was redefined as risk at any age of “giving up.”

Pattern Recognition in Older Adults’ Activities of Daily Living

Monitoring daily activities and behaviors is essential for improving quality of life in elderly care, where early detection of behavioral anomalies can lead to timely interventions and enhanced well-being. However, monitoring systems often struggle with scalability, high rates of false positives and negatives, and lack of interpretability in understanding anomalies within collected data. Addressing these limitations requires an adaptable, accurate solution to detect patterns and reliably identify outliers in elderly behavior data. This work aims to design a scalable monitoring system that identifies patterns and anomalies in elderly activity data while prioritizing interpretability to make well-informed decisions. The proposed system employs pattern recognition to detect and analyze outliers in behavior analysis, incorporating a service worker generated with Crontab Guru for automated data gathering and organization. Validation is conducted through statistical measures such as accumulated values, percentiles, and probability analyses to minimize false detections and ensure reliable performance. Experimental results indicate the system achieves high accuracy, with an occupancy probability across compartments and fewer outliers detected. The system demonstrates effective scalability and robust anomaly detection. By combining pattern recognition with a focus on interpretability, the proposed system provides actionable insights into elderly activity patterns and behaviors. This approach enhances the well-being of older adults, offering caregivers reliable information to support timely interventions and improve overall quality of life.

Outlier Detection and Correction in Smart Grid Energy Demand Data Using Sparse Autoencoders

The implementation of smart grids introduces complexities where data quality issues, particularly outliers, pose significant challenges to accurate data analysis. This work develops an integrated methodology for the detection and correction of outliers in energy demand data, based on Artificial Neural Network autoencoders. The proposed approach is submitted across multiple scenarios using real-world data from a substation, where the influence of the variation in the number of outliers present in the database is evaluated, as well as the variation in their amplitudes on the functioning of the algorithms. The results provide an overview of the operation as well as demonstrate the effectiveness of the proposed methodology that manages to improve some indices achieved by previous works, reaching accuracy and F-score superior to 99% and 97%, respectively, for the detection algorithm, as well as a square root mean squared error (RMSE) and a mean absolute percentage error (MAPE) of less than 0.2 MW and 2%, respectively.

Matrix-based long-term traffic flow prediction

ABSTRACT Accurate long-term traffic prediction is crucial for enhancing traffic efficiency, ensuring urban safety, and fostering sustainable urban development. However, forecasting over extended periods is challenging due to intricate trends, cyclical variations, and interference from outlier data. To address these issues, this study proposes a matrix-based traffic flow prediction model. The model constructs a matrix with periods as rows and similarities as columns, leveraging periodicity and similarity in traffic data. A row-column prediction module links these patterns efficiently, while a fluctuation transformation mitigates the impact of outliers, significantly improving forecast accuracy. Extending the forecast time span to 14 days with hourly intervals, the model was validated using the PeMS dataset provided by the California Department of Transportation. Results demonstrate the model’s effectiveness in capturing complex temporal dynamics, providing a robust tool for long-term traffic prediction.

Hybrid mechanism‐data‐driven iron loss modelling for permanent magnet synchronous motors considering multiphysics coupling effects

AbstractThe precise calculation of iron losses in permanent magnet synchronous motors (PMSMs) remains challenging due to the interplay between various disciplines such as electromagnetism, magnetism, and thermal/mechanical dynamics. Purely mechanistic models require detailed theoretical knowledge and exact parameters, often struggling to accurately describe complex systems, while purely data‐driven methods lack interpretability, which are susceptible to data noise and outliers in feature extraction and complicated pattern recognition. Consequently, this paper aims to present a hybrid mechanism‐data‐driven model for accurately estimating the iron loss for PMSMs, considering the multiphysics coupling effects. Specifically, based on the well‐defined physical principles, an advanced iron loss analytical model that simultaneously considers mechanical stress, temperature rise, harmonics, load currents, and changing frequency is developed and then utilised to calculate numerous loss data under different operating conditions, providing a certain level of stability and reliability for prediction accuracy. Subsequently, a convolutional neural network (CNN) algorithm is employed to perform deep learning to extract features and patterns from the data. By defining a suitable loss function, the pre‐trained model was fine‐tuned and optimised using a small amount of actual data. To validate its superiority, extensive numerical and experimental analyses are conducted on the prototype. The results demonstrate that the iron losses computed using this hybrid model overcome the limitations of singular methods by effectively leveraging both theoretical knowledge and real‐world data, thus accurately accommodating various application scenarios. This integrated approach enhances the accuracy, stability, and interpretability of the model, laying a solid foundation for more specialised applications in the future.

Chlorophyll-a Outliers in the Banda Sea and its surroundings: Implications for Ecosystem Dynamics

Chlorophyll-a concentration is a critical indicator of marine ecosystem health, reflecting primary productivity and potential ecological changes. However, the occurrence of outlier data in the Banda Sea and its surroundings has not been sufficiently studied. This study aims to investigate the dynamics of chlorophyll-a concentration outliers in the Banda Sea, focusing on their spatial and temporal patterns from 2010 to 2022. Data were acquired from MODIS satellite imagery and analyzed using R Studio for time series decomposition, while spatial patterns were explored with ArcGIS 10.8 and Google Maps to pinpoint key hotspots of outlier activity. The findings reveal fluctuating chlorophyll-a outliers with extreme concentrations ranging from 14.39 to 81 mg·m⁻³. Temporally, these outliers are predominantly observed in the western Banda Sea during January to March and December, while in the eastern Banda Sea they occur from June to September. Spatially, these anomalies are concentrated in areas of significant human activity, particularly in Tolo Bay, suggesting a potential link between anthropogenic influences and ecological disruptions. Field monitoring in areas with persistent outliers is strongly recommended to verify actual conditions and assess their ecological impacts. While chlorophyll-a plays a vital role in marine productivity, excessive concentrations can disrupt ecosystems, potentially leading to harmful algal blooms or hypoxic conditions. This study underscores the importance of integrating remote sensing, field validation, and spatial analysis to effectively monitor and manage these anomalies. Enhanced understanding of these dynamics is critical for informed decision-making and sustainable marine management practices in the Banda Sea.

A comprehensive guide to volatolomics data analysis.

Volatolomics (or volatilomics), the study of volatile organic compounds, has emerged as a significant branch of metabolomics due to its potential for non-invasive diagnostics and disease monitoring. However, the analysis of high-resolution data from mass spectrometry and gas sensor array-based instruments remains challenging. The careful consideration of experimental design, data collection, and processing strategies is essential to enhance the quality of results obtained from subsequent analyses. This comprehensive guide provides an in-depth exploration of volatolomics data analysis, highlighting the essential steps, such as data cleaning, pretreatment, and the application of statistical and machine learning techniques, including dimensionality reduction, clustering, classification, and variable selection. The choice of these methodologies, along with data handling practices, such as missing data imputation, outlier detection, model validation, and data integration, is crucial for identifying meaningful metabolites and drawing accurate diagnostic conclusions. By offering researchers the tools and knowledge to navigate the complexities of volatolomics data analysis, this guide emphasizes the importance of understanding the strengths and limitations of each method. Such informed decision-making enhances the reliability of findings, ultimately advancing the field and improving the understanding of metabolic processes in health and disease.

A lithium-ion battery SOH estimation method based on temporal pattern attention mechanism and CNN-LSTM model

Accurately estimating the state of health (SOH) of lithium-ion batteries is crucial for ensuring the availability and efficiency of the energy storage systems based on lithium-ion batteries. This paper proposes a method based on a temporal pattern attention (TPA) mechanism and a CNN-LSTM model to improve the SOH estimation accuracy. Firstly, the health factors (HFs) related to capacity degradation are selected based on the charge–discharge curves of lithium-ion batteries, and the local outliers in the health factor data are detected through the local outlier factor (LOF) method. The Lagrange interpolation is then applied to correct these outliers to ensure the continuity of the health factor data. Secondly, the effectiveness of the health factor is evaluated by using the Pearson and Spearman correlation analyses, with the valid health factor forming the HFs dataset. Thirdly, the TPA mechanism is integrated into the CNN-LSTM model to form a CNN-LSTM-TPA model to enhance its ability to capture key information. The whale optimization algorithm (WOA) is employed to optimize the model’s hyperparameters. Finally, the proposed method is tested on the NASA and CALCE lithium-ion battery datasets. The experimental results show that on the NASA dataset, the proposed method improves the SOH estimation accuracy by approximately 26.39% to 46.05% compared to the LSTM method; on the CALCE dataset, the accuracy improves by approximately 56.05% to 73.32% compared to the LSTM method, respectively. These experimental results indicate that the proposed method achieves higher accuracy.

Machine Learning-Based Outlier Detection for Business Intelligence: A Scalable Time Series Analysis Framework

The exponential growth in digital business operations has resulted in an unprecedented volume of time series data generated from diverse business metrics, creating an urgent need for sophisticated anomaly detection systems. This paper presents a comprehensive framework for detecting outliers in business time series data using advanced machine learning techniques, addressing the challenges of scale, accuracy, and real-time processing. We propose a novel hybrid approach that seamlessly integrates statistical methods with deep learning architectures to identify both point anomalies and pattern deviations in multivariate business metrics. The framework incorporates adaptive thresholding mechanisms and contextual awareness, leveraging business domain knowledge to reduce false positives while maintaining high detection accuracy across varying business cycles and seasonal patterns. Our approach addresses the challenges of scalability and real-time processing through a sophisticated distributed computing architecture, making it suitable for enterprise-scale deployments. The framework demonstrates superior performance in handling concept drift, seasonal variations, and complex interdependencies between metrics, while maintaining computational efficiency and interpretability. Keywords—anomaly detection, business intelligence, machine learning, time series analysis, deep learning, statistical methods, distributed computing, real-time processing, adaptive thresholding, feature engineering

Modeling of Gross Domestic Product Growth in Indonesia by Using Multi-Input Intervention Model

The Gross Domestic Product (GDP) Growth of Indonesia has fluctuated over time due to established policies, economic crises, changes in political direction, and natural disasters. In 1998, due to the fall of the New Order regime, the Indonesian economy contracted by -13.13 percent, leading to hyperinflation. In 2020 the COVID-19 pandemic occurred which caused Indonesia's GDP Growth to contract again. Accurate forecasting of GDP Growth is crucial for government to formulate effective future policy strategies to maintain the stability of Indonesia's economy. There are several outliers in Indonesia's GDP Growth data, so the proper analysis is a multi-input intervention. The best model analysis is ARIMA (1,0,0) with non-zero mean using the first order intervention b=0, r=0, s=0 and the second order intervention b=0, r=0, and s=0 which resulted in a Mean Absolute Percentage Error (MAPE) of 23.47 percent. The outlier effect on Indonesia's GDP Growth data is both direct and temporary.

Pengaruh CSR Terhadap Financial Performance pada Perusahaan Sub Sektor Makanan dan Minuman Tedaftar di BEI Tahun 2022-2023

This research aims to test and analyze the influence of CSR on the financial performance of manufacturing companies listed on the BEI in 2022-2023. The population of this research is all food and beverage sub-sector manufacturing companies listed on the IDX. The sample selection in this study used purposive sampling with certain criteria so that 25 companies were obtained for 2 years of sample data for research. With 11 outlier data. So the sample data obtained is 39 data. The data used is secondary data obtained from financial reports. The data analysis technique uses simple linear regression analysis. The research results show that CSR has no effect on the financial performance of manufacturing companies in the food and beverage sub-sector in 2022-2023