cover
Contact Name
Soraya
Contact Email
jurnal.varian@stmikbumigora.ac.id
Phone
+6282339979545
Journal Mail Official
jurnal.varian@stmikbumigora.ac.id
Editorial Address
Jln. Ismail Marzuki - Cilinaya - Cakranegara - Mataram 83127
Location
Kota mataram,
Nusa tenggara barat
INDONESIA
Jurnal Varian
Published by Universitas Bumigora
ISSN : -     EISSN : 25812017     DOI : https://doi.org/10.30812/varian
Jurnal Varian adalah salah satu Jurnal Ilmiah yang terdapat di Universitas Bumigora. Jurnal ini bertujuan untuk memberikan wadah atau sarana publikasi bagi para dosen, peneliti dan praktisi baik di lingkungan internal maupun eksternal Universitas Bumigora Mataram. Jurnal ini terbit 2 (dua) kali dalam 1 tahun pada periode Genap (April) dan Ganjil (Oktober). Jurnal Varian fokus memuat publikasi pada Bidang Matematika dan Statistika.
Articles 168 Documents
Analysis of Gold Price Forecasts Using Automatic Clustering Method and Fuzzy Logic Relationship Jannah, Ro'i Khatul; Agustina, Dina
Jurnal Varian Vol. 8 No. 2 (2025)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/varian.v8i2.4382

Abstract

Gold is often chosen as an investment due to its lucrative potential. To maximize profits and avoid losses, investors need to understand the volatile price movements of gold. This research aims to forecast the price of gold in the next period. In this research, the forecasting method used is Automatic Clustering and Fuzzy Logical Relationship (ACFLR). ACFLR is a method that uses the concept of fuzzy logic for modeling time series data. The forecasting process includes data sorting, cluster formation, interval determination, fuzzification, FLR and FLRG formation, and calculation of forecasting values. Based on this method, the result of the gold price forecast in Padang City for the next period, namely January 2024 using the ACFLR method is IDR 978,796.9. with a MAPE value of 0.9%, which means this method is very good. For further researchers, it is hoped that the Fuzzy Time Series method can use other forecasting models in order to obtain the most optimal method for forecasting gold prices.
Panel Data Regression Modeling with Weighted Least Squares Method Using Fair Weights Ferdiansyah, Muhammad; Raupong, Raupong; Siswanto, Siswanto
Jurnal Varian Vol. 8 No. 2 (2025)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/varian.v8i2.4392

Abstract

Panel data regression is a robust method for analyzing relationships between dependent and independent variables by combining time-series and cross-sectional data. Its reliability hinges on key assumptions, particularly homoscedasticity. Violations, known as heteroscedasticity, lead to inefficient estimates and biased inference, as estimators fail to meet the Best Linear Unbiased Estimator criteria. The Weighted Least Squares (WLS) method addresses heteroscedasticity by weighting observations based on the inverse of their variance. WLS assumes prior knowledge of the heteroscedasticity structure, which is often impractical, creating gap in evaluating its effectiveness compared to alternative methods. The purpose of this study is to examines life expectancy in South Sulawesi as the dependent variable, with expected years of schooling, per capita expenditure, and average years of schooling as independent variables. The research methode used WLS with reasonable weighting, successfully addressing heteroscedasticity. The fixed-effects model was identified as the most appropriate, with an R-squared of 99.45%. Life expectancy was explained by the model. Results shows all variables positively and significantly influence life expectancy. In conclusion, the WLS method effectively overcomes heteroscedasticity in panel data regression, providing reliable estimators. This study highlights the importance of method selection in panel data analysis and offers insights for policymakers aiming to improve life expectancy in South Sulawesi.
Evaluating Different K Values in K-Fold Cross Validation for Binary Logistic Regression to Classify Poverty Sinaga, Julia Oriana; Fathurahman, M.; Wahyuningsih, Sri; Hayati, Memi Nor
Jurnal Varian Vol. 8 No. 2 (2025)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/varian.v8i2.4403

Abstract

Data mining is essential for decision-makers to analyze and extract insights from data efficiently. Classification is one of the data mining techniques used to organize data based on its features, helping to identify patterns and make predictions. This study evaluates Binary Logistic Regression (BLR), a type of generalized linear model that suitable for binary outcomes, for classifying poverty depth across Indonesian regencies/cities in 2022, with a focus on the impact of different K values in K-Fold Cross Validation. The dataset includes 514 regencies/cities, with the Poverty Depth Index as the target variable, categorized into high (1) and low (0) levels, using 11 predictor variables. K-Fold Cross Validation was performed with K values of 3, 5, and 10, using accuracy and Area Under Curve (AUC) as evaluation metrics. The mean accuracy values for BLR are 75.7% for K=3, 74.3% for K=5, and 75.1% for K=10. Results show that K=3 offers the highest accuracy in classifying poverty depth in Indonesia, with the lowest standard deviation of 0.03. However, K=10 demonstrates superior discriminative ability in BLR, reflected by a higher AUC value. This study highlights the significant influence of K values in K-Fold Cross Validation on BLR performance.
Mengeksplorasi Masalah Kejahatan dari POV Statistik dengan Regresi Binomial Negatif Dani, Andrea Tri Rian; Fathurahman, M.; Ni'matuzzahroh, Ludia; Putri Permata, Regita; Putra, Fachrian Bimantoro
Jurnal Varian Vol. 8 No. 2 (2025)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/varian.v8i2.4445

Abstract

Criminality is a complex issue in Indonesia that is very important to the government, law enforcement agencies, and society. The underlying causes of Indonesia's crime problem are complex and impacted by various circumstances. The aim of this research is to model the crime problem in Indonesia and determine the influencing factors.  The method used in this research is Negative Binomial Regression. The results of the study show that the negative binomial regression model can be used to model criminal problems because the variance value is more significant than the average. Based on the parameter significance test results, both simultaneously and partially, the open unemployment rate, Gini ratio, average years of schooling, and prevalence of inadequate food consumption significantly affect the crime rate, with an Akaike’s Information Criterion Corrected (AICc) value of 698,098. These findings suggest that addressing economic inequality, unemployment, education, and food security could help reduce crime in Indonesia. Policies aimed at improving job opportunities, reducing income disparity, and enhancing education and food security are crucial in mitigating crime. This study provides valuable insights for policymakers and law enforcement agencies, offering a foundation for more targeted and effective crime prevention strategies. Future research could employ the robust Poisson Inverse Gaussian Regression method to avoid the overdispersion problem. 
Simulating the Dynamics of Early Marriage and Marital Stability Using SERH Mathematical Models Khairana, Nadiyah; Annas, Suwardi; Side, Syafruddin; Sainon Andi Pandjajangi, Andi Muhammad Ridho Yusuf
Jurnal Varian Vol. 8 No. 2 (2025)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/varian.v8i2.4460

Abstract

This study aims to develop a mathematical model of SERH (Susceptible, Engaged, Risk, and Stabilization) to analyze and predict the incidence of early marriage in South Sulawesi Province. The research employs method a combination of theoretical and applied approaches, utilizing differential equations to model the dynamics of early marriage spread. Data were collected through questionnaires distributed to 400 couples selected using the Slovin technique, representing a population of 57,789 couples. The SERH model parameters, including the rate of couple interaction , transition from engagement to risk , and recovery from risk to stability , were derived from the collected data. Simulations were conducted using Maple software to visualize the spread of early marriage under different scenarios. The results of the analysis revealed two equilibrium points: a marriage-free equilibrium and a stable endemic equilibrium. The basic reproduction number  was calculated to be 3.97, indicating that one couple can influence 3-4 others in their social environment. However, with effective interventions such as education and counseling, the R₀ value can be reduced to 0.45, significantly lowering the spread of early marriage. This study provides valuable insights for policymakers to design targeted prevention programs and highlights the importance of early intervention in reducing the prevalence of early marriage.  
Multiple Regression Model on Selling Price, Sales Volume, Raw Material Costs, and Direct Labor Costs on Profit Machfiroh, Ines Saraswati; Maulida, Afna
Jurnal Varian Vol. 8 No. 2 (2025)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/mqvzph91

Abstract

During its operations, a company always aims to achieve increased profits. The case of PT Ciomas Adisatwa RPA Unit Banjarmasin from 2019 to 2022 demonstrates a trend of rising profits. Several factors can influence a business's profitability, including selling price, sales volume, and incurred costs. The objective of this study is to analyze how the profit of PT Ciomas Adisatwa RPA Unit Banjarmasin is influenced by selling price, sales volume, raw material costs, and direct labor costs. This research applies a quantitative descriptive approach using secondary data. The monthly data utilized spans from 2019 to 2022, comprising a total of 48 data points. The results of the study show that selling price, sales volume, and raw material costs have a significant effect on the profit of PT Ciomas Adisatwa RPA Unit Banjarmasin, while direct labor costs do not have a significant influence. These findings imply that pricing strategies and increased sales volume are the main factors that can enhance the company's profitability. Additionally, controlling raw material costs is a crucial aspect that must be managed efficiently to avoid burdening the cost structure. On the other hand, direct labor costs can be minimized through production efficiency and the use of technology without compromising production output. Therefore, the company should focus on pricing strategies, marketing, and raw material cost efficiency as key measures to sustainably increase profits.  
Rainfall Forecasting Using the Singular Spectrum Analysis (SSA) Method Nurhikmawati, Nurhikmawati; Aswi, Aswi; Ahmar, Ansari Saleh
Jurnal Varian Vol. 8 No. 2 (2025)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/varian.v8i2.4571

Abstract

This study aims to evaluate the accuracy and performance of rainfall data forecasting in the city of Parepare using the Singular Spectrum Analysis (SSA) method. Situated in South Sulawesi Province, Parepare City is characterized by high rainfall intensity, which increases the likelihood of natural hazards such as flooding and landslides. These disasters have the potential to negatively impact key sectors, including economic activity, tourism, and transportation. Therefore, reliable rainfall prediction plays a crucial role in establishing a robust disaster early warning system. Monthly rainfall measurements from two stations, Bukit Harapan and Bulu Dua, are analyzed. The results reveal a Root Mean Square Error (RMSE) of 191.0566 for Bukit Harapan station and 346.023 for Bulu Dua station, underscoring the method's forecasting accuracy. A 12-month forecast predicts consistently high monthly rainfall in Parepare City, with the highest rainfall expected in December 2024 at Bukit Harapan station and in January 2024 at Bulu Dua station. Conversely, the lowest rainfall at both stations is anticipated in July 2024. Forecasts predicting increased rainfall during certain periods, especially in December and January, provide critical insights for strengthening disaster preparedness and informing mitigation strategies. This information also plays a key role in minimizing adverse effects on the economic, transportation, and tourism sectors, while promoting more efficient and sustainable management of water resources.  
Estimating and Forecasting Composite Index in Pandemic Era Using ARIMA-GARCH Model Hidayat, Agus Sofian Eka; Primajati, Gilang
Jurnal Varian Vol. 7 No. 2 (2024)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/varian.v7i2.2103

Abstract

Many industries have suffered financial losses as a result of the COVID-19 epidemic. The stock market's movement has been impacted by this circumstance. Due to the influence of some people, a large number of individuals with limited trading knowledge are attempting to participate in the stock market. Market volatility should be understandable in order to gain profit instead of having losses. Therefore, it's essential to comprehend the market of the future through analysis of the data. The purpose of this study is to use ARIMA-GARCH to predict the Indonesian stock market price during. The period covered by the dataset is January 2020–December 2022. The training data indicates that ARIMA (2,1,2) is the best model for ARIMA. The results showed that data residual fitted by ARIMA (2,1,2)-GARCH (1,2) exhibits heteroscedasticity, according to the residual analysis. The MAPE score is 2%, which is relatively small. It means that ARIMA (2,1,2)-GARCH (1,2) is good enough for forecasting the Jakarta Composite Index.
Naive Bayes Algorithm with Feature Selection Using Particle Swarm Optimization Siswanto, Siswanto; Kurniawan, Iwan; Thamrin, Sri Astuti
Jurnal Varian Vol. 7 No. 2 (2024)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/varian.v7i2.2409

Abstract

The COVID-19 vaccine in Indonesia has led to the emergence of public opinion which is conveyed on social media such as Twitter. One of the analyses that can be done to produce various information from public opinion is sentiment analysis. Sentiment analysis is used to determine whether an opinion tends to be positive or negative. This study aims to classify the public opinion of the COVID-19 vaccine in Indonesia with sentiment analysis and to visualize the location of the sentiment of the COVID-19 vaccine tweet data in Indonesia. To achieve this aim, the Naïve Bayes algorithm with Particle Swarm Optimization (PSO) feature selection was used. This study uses opinions into positive and negative class sentiments towards 2,547 tweets related to the COVID-19 vaccine in Indonesia from January to June 2021. The results show that the distribution of positive and negative class sentiments is 2,328 and 219, respectively. In addition, the positive sentiment for the COVID-19 vaccine was dominated by people on the island of Java based on a random number matrix initialized by the PSO method. The classification of public opinion on Twitter media provides accurate and optimal performance results using a combination of the Naïve Bayes algorithm with PSO feature selection. The results of the combination of these methods have accuracy and F1 score values of 91.28% and 95.38%, respectively. The visualization of geo-spatial mapping showed that positive sentiments related to the COVID-19 vaccine exist in almost all regions in Indonesia but are dominated by the Jabodetabek area.
Improved Chi Square Automatic Interaction Detection on Students Discontinuation to Secondary School Al Anshory, Fadhil; Siswanto, Siswanto; Thamrin, Sri Astuti; Inayah, Ika
Jurnal Varian Vol. 7 No. 1 (2023)
Publisher : Universitas Bumigora

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30812/varian.v7i1.2627

Abstract

Improved Chi Square Automatic Interaction Detection (CHAID) with bias correction is the development of the CHAID method by relying on Tschuprow's T test calculations with bias correction in the process of forming a classification tree. This study aims to obtain a classification of factors which influence students for not continuing their education from junior high school or equivalent to high school or equivalent. The results obtained in the classification tree produce nine classifications. Based on the results of the classification tree, the classification of students who do not continue their education to high school or equivalent is: students with disabilities who do not have access to Information and Communication Technology (ICTs) (0.89); students who work without disability but do not have access to ICTs (0.73); and students who do not work without disability but do not have access to in ICTs (0.60). Based on the classification obtained the factors which influence students for not continuing their education to high school or equivalent are access to ICTs, employment status, and persons with disabilities. The classification accuracy of the results uses the Improved-CHAID method with bias correction with a proportion of 80% training data and 20% testing data, namely 72.3033% on training data and an increase of 73.3300% on testing data.