Journal Of Data Insights

Prediction of Covid-19 Cases in Indonesia Using the Auto Regressive Integrated Moving Average Method

Asriyanti Sawiah Adam — 2025-06-30

This study discusses the use of the ARIMA (Auto Regressive Integrated Moving Average) model to predict the number of COVID-19 cases in Indonesia based on previous data. The results of the analysis show that the ARIMA (1,0,0) model is the most accurate in predicting the spread of COVID-19. Based on this model, the prediction results obtained that confirmed COVID-19 data from January to December 2022 are predicted to decrease. The number of confirmed cases of COVID-19 until December 2022 is predicted to reach 20,0365 cases of spread. So this Covid-19 case still needs special and more serious attention from the government and the public must still be vigilant because based on the results of the study there have been no signs of a significant decrease in the spread of Covid-19 cases. This study provides important insights for the government, medical personnel, and the public in planning strategies for preventing and handling the pandemic

Panel Data Regression Approach to Identify Factors Affecting Unemployment in East Java Province

Rizka Amalia Putri — 2025-06-30

The Open Unemployment Rate (OOP) in East Java Province is a multidimensional problem influenced by economic and social factors, with significant disparities between districts/cities. This study analyses the effect of Poverty Percentage, Labour Force Participation Rate (TPAK), and Economic Growth on the open unemployment rate using a panel data regression approach to accommodate spatial and temporal heterogeneity. Cross-section (38 districts/cities) and time series (2019-2021) data were analysed through three models: Common Effect Model (CEM), Fixed Effect Model (FEM), and Random Effect Model (REM). The results of statistical tests (Chow, Hausman, and Lagrange Multiplier) show the FEM as the best model with a coefficient of determination of 0.555, explaining 55.5% of the variation in the unemployment rate. The FEM estimation reveals that the Poverty Percentage has a significant positive effect on increasing the unemployment rate, while Economic Growth has a negative impact on reducing the unemployment rate. This finding confirms the need for policies focused on poverty alleviation and increasing economic growth based on regional leading sectors. This study enriches the methodological literature through the application of FEM that controls for region-specific heterogeneity, while providing practical recommendations for policy makers in designing precise unemployment reduction interventions, such as skills training based on industry needs and strengthening labour-intensive programmes.

Forecasting Honda Car Retail Sales Using the Seasonal Autoregressive Integrated Moving Average Method

Lea Angelina — 2025-06-30

This article discusses the forecasting of Honda car retail sales using the Seasonal Autoregressive Integrated Moving Average (SARIMA) method. The study aims to forecast Honda car retail sales for the upcoming year. Various SARIMA models have been tested to determine the best model, and the results show that the SARIMA (1,1,0)(1,1,1)¹² model provides the lowest Mean Absolute Percentage Error (MAPE) among all tested models, which is 17,74%. Therefore, this model was chosen for forecasting sales over the next 12 months. The forecast results are expected to assist management in making optimal decisions regarding stock and marketing, as well as significantly enhancing operational efficiency and customer satisfaction in the future.

Evaluation of Deep Learning Optimizers for Predicting JISDOR Exchange Rates Using LSTM Networks

Ariska Fitriyana Ningrum — 2025-06-30

This research explores the application of four optimization algorithms—Adam, Nadam, RMSProp, and SGD—on a Long Short-Term Memory (LSTM) model to forecast the Jakarta Interbank Spot Dollar Rate (JISDOR). The volatile nature of exchange rate data, influenced by global and domestic economic dynamics, necessitates the use of models like LSTM that excel in capturing both short- and long-term dependencies. Performance was assessed using metrics such as Root Mean Squared Error (RMSE) and Mean Absolute Percentage Error (MAPE). Among the optimizers, Nadam proved to be the most effective, achieving the lowest RMSE of 62.767 and a MAPE of 0.003, indicating its capability in managing complex fluctuations in the dataset. Despite Nadam's promising results, opportunities for improvement remain, including the inclusion of additional input variables, fine-tuning model parameters, and expanding the training dataset. This study underscores the critical role of selecting appropriate optimization algorithms for enhancing the accuracy of LSTM models in forecasting volatile financial time-series data, particularly for currency exchange rates

Implementation of K-Means Algorithm to Group Age of Cardiovascular Disease Patients

Mulya Asy-syifa Rahmi — 2025-06-30

Cardiovascular disease, including coronary heart disease, peripheral arteries and heart failure, is a serious disease that is the leading cause of death globally. Risk factors such as high blood pressure, dyslipidemia, smoking, diabetes, and obesity contribute to the development of this disease. This study aims to group cardiovascular disease sufferers based on age using the k-means clustering method with optimization of the k value using the elbow method. The data used comes from more than 35,000 preprocessed observations. The analysis results show that the optimal number of clusters is five. Data preprocessing succeeded in cleaning the data from missing values, and the elbow method helped determine the number of clusters that were relevant for age grouping of cardiovascular disease sufferers. The results of this grouping can be used for further analysis in efforts to prevent and manage cardiovascular disease.

Implementation of Hierarchical Clustering for Grouping Economic Development Indicators in Central Java Province

Salmaa — 2025-06-30

In the midst of global economic shifts, the economy in Indonesia must continue to improve. To help economic recovery after the contraction caused by the COVID-19 pandemic, the Indonesian government has implemented various policies. One way is through the process of increasing per capita income over a long period of time, known as economic development, provided that the number of people living below the absolute poverty line does not increase and income distribution does not decrease. Other efforts can be made by analyzing economic development indicators. One method that can be used is hierarchical cluster analysis to group economic development indicators in Central Java province. Average linkage is used as an approach method after carrying out correlation analysis of the five approaches in hierarchical analysis because the correlation value is the highest. From this analysis two clusters were produced with the first cluster having higher characteristic values compared to the second cluster.

K-Nearest Neighbor (KNN) Method for Weather Data Prediction

Agata Dwi Putri Putri — 2025-06-30

The weather tends to change frequently every day, so weather forecasts are made to be used as an early warning if sudden weather changes occur. By forecasting the weather, losses can be minimized and people are alert to carry out outdoor activities. From this problem, the K-Nearest Neighbor (KNN) method was applied. This method is expected to provide accurate and efficient information to obtain weather predictions for existing conditions. The data used is secondary data. After conducting research on training data (old data) amounting to 80% and test data (new data) amounting to 20%. The accuracy results from the testing data predictions are 75% with a value of k = 8.