J Statistika: Jurnal Ilmiah Teori dan Aplikasi Statistika

Binary Logistic Regression Analysis on the Spread of Dengue Fever in Bali Province

Luh Putu Safitri Pratiwi — 2024-07-31

The spread of dengue fever involves a complex cycle between humans as hosts and mosquitoes as vectors. Symptoms of dengue fever can vary from a mild fever to a severe form that can be life-threatening. One of the areas that has the highest spread of dengue fever in Bali Province is the Denpasar area. Research continues to be carried out to understand the factors that influence the spread of dengue fever using the binary logistic regression method. Binary logistic regression is a regression model that is often used in modeling categorical data, where the dependent variable in this study is the distribution of dengue fever cases with the number of cases spreading in each region being assigned a category of zero for a low number of cases and one for a high number of cases. So in this research a more effective strategy was developed in controlling this disease as well as the best model for data on the spread of dengue fever in Bali Province. The results obtained from this research were a test of the influence of the independent variable on the dependent variable, showing that the variable number of adequate sanitation facilities (X₅) had a significant influence on the number of dengue fever sufferers in Bali Province, namely 0.081.

Association of Poverty Categories, Educational Characteristics, and Area of Residence in Indonesia Using a Three-Way Log-Linear Model

Ainun Salsabila — 2024-07-31

Contingency tables are one way to present data with all categorical variables. The analysis used to model the contingency table is a log-linear model. The log-linear model is also used to estimate parameters and see the association between variables. This research aims to utilize the three-way log-linear model to model and see the association between poverty category variables, educational characteristics (level of education and reading and writing ability) of the head of the household, and area of residence in Indonesia in 2023. Research is done by forming a saturated and homogeneous log-linear model first, then comparing the difference in deviance values from the two models with the table chi-square value or choosing the smallest AIC value to determine the best model. The results obtained are a significant saturated model. This means that there is an association between the poverty category variable, the education level of the head of the household, and the area of residence. There is also an association between the poverty category variable, the reading and writing ability of the head of the household, and the area of residence. In addition, there is a greater tendency for poverty for heads of households who have a primary school education or less and cannot read and write.

Factors Affecting the Resilience Index Food in Papua Province and West Papua Province Using a Spatial Model Approach

Dewi Sri Hastuti — 2024-07-31

The Food Security Index is a measure of indicators to produce a composite value that reflects the status of food security in a region. Food security plays an important role in sustainable development, including food availability, environmental preservation and economic balance, as well as being the basis for economic growth, preventing poverty and inequality. In Indonesia, with an estimated population growth of 430 million people in 2050, the challenge of meeting food needs is increasing. Indonesia's commitment to the Sustainable Development Goals (SDGs) includes efforts to end hunger and promote sustainable agriculture. This research aims to apply spatial regression analysis to the Provinces of Papua and West Papua to determine the best model and significant factors that influence the Food Security Index in the region in order to identify the challenges faced by the region in calculating the food availability of its people as well as assist in developing efforts to overcome them. Five predictor variables were used with the assumption that they have a significant influence on the Food Security Index. This research examines the spatial regression equation using the SAR, SEM and SARMA regional approaches. The results obtained showed that the selected SEM model with a p-value of 0.0082581 was appropriate for identifying the dependence of spatial effects on Food Security Index in Papua Province and West Papua Province. Life Expectancy at Birth, Prevalence of Stunting Toddlers, Percentage of Poor Population, Open Unemployment Rate, and Average Length of Life are significant factors that influence the Food Security Index in Papua Province and West Papua spatially.

Sentiment Analysis Of Public Opinion On Handling Stunting In Indonesia Using Random Forest

Ariska Fitriyana Ningrum — 2024-07-31

The problem of stunting is important to solve, as it has the potential to disrupt human resource potential and is linked to health outcomes and even child mortality. The Indonesian government targets the stunting rate to drop to 14 percent by 2024 through an accelerated stunting reduction program as an effort to improve the nutritional status of the community and also reduce the prevalence of stunting or short toddlers. Understanding public sentiment towards stunting initiatives is essential for policy makers and stakeholders to design effective interventions and allocate resources efficiently. In this research, classification of positive and negative sentiment is carried out using the random forest algorithm. The data used is comment data on one of the social media pages, namely Twitter, regarding public sentiment towards handling stunting cases in Indonesia. The first stage in this research after obtaining a data is data preprocessing. The data preprocessing stage in sentiment analysis is useful for cleaning and normalizing text, removing irrelevant words, and preparing data so that algorithms can analyze sentiment more accurately and efficiently. Furthermore, the results of the preprocessed data are labeled 0 for positive and 1 for negative labels. The classification of positive and negative sentiment was done using random forest and resulted in an accuracy value of 97.5%. This model is good, but we suggest trying other algorithms in future research.

Accelerating SIREKAP Digital Transformation in the 2020 Natuna Regency Election

Bagus Aji Kuncoro — 2024-07-31

The aim of this research is to analyze the implementation of digital transformation in the use of Sirekap in Pilkada 2020 district of Natuna. This type of research is qualitative, using a single case study research strategy, which involves an individual in one company or office, namely the Natuna Regency KPU office. The research design used is qualitative research. The acceleration of digital transformation is essential in the operational process of acquisition and as a support tool while minimizing risks from the early stages of the elections, the election day, and the post-election period. (post-election). The over-implementation of the Sirekap in the election of the head of the Bupati district and the Deputy Bupati District of Natuna district in 2022 increases transparency and accountability to increase public confidence in the results of the election calculations. The Sirekap application makes the working time of the KPU more effective than manual calculations. The KPU, with the presence of the Sireap application, also makes the information disseminated to the public no less rapidly than the survey agency because the region can monitor the data entered in the place of direct voting. The Sirekap application also has a high level of rigor and minimizes the error rate of voting calculation

Forecasting Average Rice Prices at Milling Level According to Quality Using Support Vector Regression

Fauziah Roshafara — 2024-07-31

Indonesia is an agricultural country where the majority of the population work as farmers and one of the humongous commodities produced is rice. Rice is a very important commodity for the Indonesian people, because it is the main food of them. This is why rice production in Indonesia is the big concern to the government, including of the average rice prices at milling level. The fluctuative of the rice prices will be affect to the purchasing power of the people. One of the efforts that can be made to prepare a policy to increase people's purchasing power of the rice is by forecasting. This study used SVR to modeling the average rice prices using 114 datasets obtained from January 2013 to June 2023, then evaluating its performance using Mean Absoute Percetage Error (MAPE). The best model formed from a linear kernel with parameters ε = 0.078 and C = 3.1. The model produced the smallest MAPE value of 2.32% in testing data and 1.2% in training data which also less than 10% meaning that the performance of the model to forecast the average price of rice is very high.

Dynamical Analysis of Mathematical Model of Social Behavior with Law Enforcement and Religious Approaches

Nailul Izzati — 2024-07-31

Mathematical models could be used to describe various phenomena, one of which is social phenomena. One of the interesting social phenomena to study is criminal behavior. By dividing the total population into several compartments based on their social behavior status, a mathematical model could be constructed to depict social dynamics. These social dynamics could be understood through dynamical analysis and numerical simulation. In this study, numerical simulations were conducted using Maple 2022 software and the Runge-Kutta method. Based on the results of dynamical analysis and numerical simulations, it is known that by implementing law enforcement and religious approaches, criminal activities within a population could be reduced or even eliminated.

The Effect of Motivation, Work Discipline, and Organizational Commitment on Employee Performance at PT. Hebsa Indonesia

Afina Safarah — 2024-07-31

Human resources are a crucial asset in an organization, as they play a key role in implementing policies and operational activities. Companies with good human resources will achieve optimal results in their performance. It is essential for organizations to refine their planning to enhance employee productivity. This research aims to examine the impact of motivation, work discipline, and organizational commitment on employee performance at PT. Hebsa Indonesia, involving all 35 employees. The sampling technique used in this study is saturated sampling. The research method employed is quantitative. Data analysis was conducted using SmartPLS 3 software with the Partial Least Square (PLS) approach. The results indicate that motivation does not have an effect on employee performance at PT. Hebsa Indonesia, while work discipline and organizational commitment do influence employee performance.

Statistical Quality Control (SQC) Method Analysis Regarding Qualty Control of Shoe Products (Case Study of PT-X)

Dona Rizqi Aulia — 2024-07-31

The shoe industry plays an important role in the economy with mass production, creating fierce competition. Companies need to plan production carefully to meet market demand. Observations show that there are production defects that reduce performance and cause losses, especially in distribution. This research uses P-Chart Analysis to compare the proportion of defective products, aiming to detect errors before sale. The application of a quality control method through Statistical Quality Control (SQC) is proposed. The Pareto diagram identifies the main types of damage: over cementing (35.9%), open bonding (27.7%), overlay (24.1%), metal material damage (6.3%), and material damage (6.1 %). P-Chart analysis found points outside the control limits in the 15th and 18th data, which were removed for sample stabilization. Fishbone analysis helps identify the cause of the problem. Over cementing, open bonding, and overlay are caused by human, material, and machine factors. Damage to metal materials and materials is also caused by human, material and machine factors, as well as method factors. This research shows the importance of implementing quality control to reduce production defects and improve company performance.

Distribution of Soap X in East Java Region with Bhumal Method and Traveling Salesman Problem

Hery Murnawan — 2024-07-31

The product distribution process is an effort by producers to deliver their products to consumers with a registered and programmed system. Product distribution to consumers must be handled carefully and carefully. Because this can be one of the obstacles for companies in winning competition with other companies if the distribution speed is not high. The speed or slowness of product distribution to consumers, one of which depends on the proximity between the dealer's warehouse and the market (consumers). The purpose of this data presentation and analysis is to better understand how to make data, analyse, and make conclusions that can determine a decision on a logistics process in general and bar soap products with transportation cost optimisation using the bhumal method and Traveling Salesman Problem.