cover
Contact Name
Ansari Saleh Ahmar
Contact Email
jurnalvariansi@unm.ac.id
Phone
-
Journal Mail Official
jurnalvariansi@unm.ac.id
Editorial Address
Program Studi Statistika, Fakultas MIPA UNM, Jalan Daeng Tata Raya, Makassar, 90223
Location
Kota makassar,
Sulawesi selatan
INDONESIA
VARIANSI: Journal of Statistics and Its Application on Teaching and Research
ISSN : -     EISSN : 26847590     DOI : http://dx.doi.org/10.35580/variansiunm26374
VARIANSI: Journal of Statistics and Its application on Teaching and Research memuat tulisan hasil penelitian dan kajian pustaka (reviews) dalam bidang ilmu dasar ataupun terapan dan pembelajaran dari bidang Statistika dan Aplikasinya dalam pembelajaran dan riset berupa hasil penelitian dan kajian pustaka.
Articles 16 Documents
APLIKASI MULTIVARIATE ADAPTIVE REGRESSION SPLINES (MARS) UNTUK MENGETAHUI FAKTOR YANG MEMPENGARUHI CURAH HUJAN DI KOTA MAKASSAR Muhammad Reski Mattalunru; Suwardi Annas; Muhammad Kasim Aidid
VARIANSI: Journal of Statistics and Its application on Teaching and Research Vol. 4 No. 1 (2022)
Publisher : Program Studi Statistika Fakultas MIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (846.514 KB) | DOI: 10.35580/variansiunm2

Abstract

Analisis regresi nonparametrik merupakan metode alternatif ketika asumsi parametrik terlanggar. Kemampuan estimasi yang tinggi serta sifatnya yang fleksibel membuat regresi nonparametrik menjadi sebuah pemodelan masa kini dan masa mendatang. Memperhatikan gejala alam dewasa ini semakin hari semakin sulit untuk diduga. Musim hujan merupakan salah satu fenomena alam yang semakin hari semakin mengarah pada pola yang tidak menentu. Bulan yang biasanya telah menjadi penanda musim kemarau malah tiba-tiba terjadi curah hujan yang sangat deras bahkan mengakibatkan banyak kerugian. Maka dibutuhkan pemodelan untuk mengetahui faktor-faktor apa yang mempengaruhi curah hujan. Metode Multivariate Adaptive Regression Splines (MARS) merupakan salah satu metode pemodelan modern dengan kemampuan estimasi yang tinggi. Selain itu MARS memilki sifat yang fleksibel serta ketangguhan mengatasi data yang berdimensi tinggi yaitu data yang memiliki variabel bebas 3 ≤ x ≤ 20 dan ukuran data sampel 50 ≤ n ≤ 1000. Model MARS diperoleh dari kombinasi antara Basis Fungsi (BF), Maksimum Interaksi (MI) dan Minimum Observasi (MO) dengan Generalized Cross Validation (GCV) yang bernilai kecil. Pada penelitian ini banyaknya variabel bebas yang digunakan sebanyak 4 variabel. suhu udara, kelembaban udara, kecepatan angin, dan tekanan udara merupakan variabel bebas yang mempengaruhi curah hujan di Kota Makassar dengan tingkat kontribusi masing-masing sebesar 86,54%, 100%, 39,38% dan 54,68%. Kombinasi model terbaik MARS pada penelitian ini adalah BF=12, MI=1, dan MO=1 dengan GCV=31,14
ANALISIS BAYESIAN SURVIVAL WEIBULL UNTUK MENENTUKAN FAKTOR YANG MEMPENGARUHI LAJU KESEMBUHAN PASIEN RAWAT INAP KANKER SERVIKS DI RSDU KOTA MAKASSAR Nini Harnikayani Hasa; M Nadjib Bustan; Aswi Aswi
VARIANSI: Journal of Statistics and Its application on Teaching and Research Vol. 4 No. 1 (2022)
Publisher : Program Studi Statistika Fakultas MIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (501.876 KB) | DOI: 10.35580/variansiunm6

Abstract

Survival analysis is a statistical procedure for analyzing data where the response variable is the time until the occurrence of an event. In this study, Bayesian survival Weibull was used to determine the factors that influence the rate of recovery of cervical cancer inpatients. The data used in this study is cervical cancer inpatient data at the Makassar City Hospital for the 2017-2019 period. Based on the results of the analysis, it was found that a significant factor affecting the healing rate of cervical cancer inpatients was complications. Cervical cancer inpatients who experience complications tend to recover slower by 0.258 than patients who do not experience complications.
PEMODELAN LAJU INFLASI DENGAN MENGGUNAKAN REGRESI NON-LINEAR BERBASIS ALGORITMA GENETIKA (Kasus: Kota-Kota di Pulau Jawa) Wildan Mujahid; Muhammad Arif Tiro; Ruliana Ruliana
VARIANSI: Journal of Statistics and Its application on Teaching and Research Vol. 4 No. 1 (2022)
Publisher : Program Studi Statistika Fakultas MIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (381.978 KB) | DOI: 10.35580/variansiunm7

Abstract

This research is applied research that uses non-linear regression on the inflation rate data and the factors that are thought to influence it. By using the RESET Test, statistics are obtained, namely the RESET value = 3.7506 with P value = 0.04138, which means that the inflation data is appropriate to use non-linear regression. From the results of this study, it was found that the average inflation rate of 26 cities in Java was 22.08% with a standard deviation of 24.33%. From the results of this study it was also found that the consumer price index (X1), city/district minimum wages (X2), and regional gross domestic product (X3) are factors that affect the inflation rate with the best model with an RMSE value of 0.445.
ANALISIS SUPPORT VECTOR REGRESSION (SVR) DENGAN KERNEL RADIAL BASIS FUNCTION (RBF) UNTUK MEMPREDIKSI LAJU INFLASI DI INDONESIA Isnaeni R; Sudarmin Sudarmin; Zulkifli Rais
VARIANSI: Journal of Statistics and Its application on Teaching and Research Vol. 4 No. 1 (2022)
Publisher : Program Studi Statistika Fakultas MIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (977.99 KB) | DOI: 10.35580/variansiunm13

Abstract

Inflation is one indicator that affects the economic growth of a country. As a developing country, Indonesia has an unstable inflation rate every year. Therefore, it is necessary to predict the inflation rate in the future to be useful for formulating future economic policies. SVR is a Support Vector Machine (SVM) development for regression cases. In the SVR method, the RBF kernel is used as an aid in solving non-linear problems, the Min-Max Normalization method for data normalization, distribution of training data and testing data, selecting the best model with Grid Search Optimization, then forecasting using the model obtained with parameter = 0,1, C = 1, and = 3. The forecasting results obtained were evaluated by looking at the RMSE value, the test value obtained was RMSE of 0.0020, which means the model's ability to follow the data pattern well
Analisis Ridge Robust Penduga Generalized M (GM) Pada Pemodelan Kalibrasi Untuk Kadar Gula Darah Agung Tri Utomo; Erfiani Erfiani; Anwar Fitrianto
VARIANSI: Journal of Statistics and Its application on Teaching and Research Vol. 4 No. 2 (2022)
Publisher : Program Studi Statistika Fakultas MIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (384.666 KB) | DOI: 10.35580/variansiunm14

Abstract

Calibration modeling is one of the methods used to analyze the relationship between different methods. The relationship is like the relationship between invasive and non-invasive blood sugar measurement. Problems that often arise in calibration modeling are multicollinearity and outliers. Multicollinearity problems can cause the regression confidence interval to widen, so that there is no statistically significant regression coefficient. Outliers cause statistical tests to deviate. The handling of these problems can be solved by robust ridge analysis. Ridge robust is a combined analysis of ridge regression and robust regression. Ridge regression is able to overcome the problem of multicollinearity and robust regression can overcome the problem of outliers. The estimator used is Generalized M (GM). This method will be applied to a calibration model that uses invasive and non-invasive blood sugar level data. The model used with Generalized M (GM) estimator robust regression using modulation clusters 50 to 90 in 2017 is better than the modulation group 50. up to 90 in 2019. The statistical values obtained are SSE of 0.910, RMSEadj of 0.114, and RMSEP of 0.030. Calibration models that have outliers and multicollinearity problems can be overcome by robust ridge regression. The feasibility value of the model obtained in the GM estimator robust regression is smaller than the MM estimator ridge robust regression in the calibration modeling for non-invasive blood sugar level data. That is, the best model that can be used is the robust ridge regression GM estimator.
PENGEMBANGAN PAKET R UNTUK ANALISIS DISKRIMINAN BERBASIS GRAPHICAL USER INTERFACE WEB INTERAKTIF Nur Isra; Suwardi Annas; Muhammad Kasim Aidid
VARIANSI: Journal of Statistics and Its application on Teaching and Research Vol. 4 No. 3 (2022)
Publisher : Program Studi Statistika Fakultas MIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (895.351 KB) | DOI: 10.35580/variansiunm24

Abstract

Penggunaan perangkat lunak berlisensi memerlukan biaya yang relatif mahal, dan sulitnya memperoleh perangkat lunak berlisensi menjadi salah satu penyebab meningkatnya penggunaan perangkat lunak bajakan. Salah satu upaya dalam mengurangi tingkat permasalahan perangkat lunak bajakan adalah melakukan pengembangan perangkat lunak yang memiliki lisensi publik bersifat open source seperti perangkat lunak R. Penelitian ini dilakukan untuk menyusun beberapa paket yang terdapat pada perangkat lunak R yang akan memudahkan pengguna dalam melakukan analisis statistika, khususnya untuk analisis diskriminan linear. Paket pendukung R tersebut yaitu paket R-Shiny yang mampu membuat tampilan berbasis Graphical User Interface. Pengembangan paket R dalam penelitian ini menggunakan metode waterfall. Paket ini bernama Linear Discriminant Analysis Application (LDA App). Berdasarkan pengujian yang dilakukan pada LDA App menunjukkan bahwa LDA App mampu menyelesaikan analisis statistika sesuai fungsinya. Perbandingan antara LDA App dan software statistika lainnya memiliki ouput yang sama, akurat, dan lebih efisien dalam melakukan analisis diskriminan linear.
Metode Subtractive Fuzzy C-Means (SFCM) dalam Pengelompokan Kabupaten/Kota di Provinsi Sulawesi Selatan Berdasarkan Indikator Kemiskinan Nurul Imania Kalla; Suwardi Annas; Muhammad Fahmuddin
VARIANSI: Journal of Statistics and Its application on Teaching and Research Vol. 4 No. 2 (2022)
Publisher : Program Studi Statistika Fakultas MIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (868.759 KB) | DOI: 10.35580/variansiunm25

Abstract

The percentage of poor people in South Sulawesi Province has increased in the last 3 years. It is necessary to handle poverty that by adjusting to the poverty level of each district. It is necessary to have a description of socio-economic conditions by grouping regency/city based on poverty indicators. The purpose of this grouping is determining the poverty level of each district so that it can be used as material for evaluating poverty problems. The clustering method often used in research was Fuzzy C-Means (FCM). In the development, FCM was combined with the Subtractive Clustering (SC) method to obtain the hybrid Subtractive Fuzzy C-Means (SFCM) method. The SFCM method has the advantage of speed, iteration, resulting in a data partition that is more stable and accurate when it is compared to the previous method. On this research, the SFCM method was applied with 4 variables from the poverty indicator data for South Sulawesi Province in 2021. The results of the analysis show that based on the Partition Coefficient Index (PC), the percentage of poverty in South Sulawesi Province was divided into three clusters, namely the low, medium and high poverty percentage clusters. Clusters of low poverty percentage consist of Bulukumba, Bantaeng, Takalar, Gowa, Soppeng, Wajo, Sidenreng Rappang, East Luwu, Makassar City, Parepare City, and Palopo City. Clusters of medium poverty percentage consist of Sinjai, Maros, Barru, and Enrekang. Meanwhile, Clusters of high poverty percentage consist of the Selayar Islands, Jeneponto, Pangkajene and Islands, Bone, Pinrang, Luwu, Tana Toraja, North Luwu, and North Toraja
ANALISIS HIERARCHICAL CLUSTERING MULTISCALE BOOTSTRAP (KASUS: INDIKATOR KEMISKINAN DI PROVINSI SULAWESI SELATAN TAHUN 2020) Musdalifah M. Ramly; Sudarmin Sudarmin; Bobby Poerwanto
VARIANSI: Journal of Statistics and Its application on Teaching and Research Vol. 4 No. 3 (2022)
Publisher : Program Studi Statistika Fakultas MIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (805.252 KB) | DOI: 10.35580/variansiunm26

Abstract

Hierarchical cluster analysis is a statistical analysis used to group data based on their similarities. The single linkage, complete linkage and average linkage methods can be used to group data using distance techniques. There is a large difference in the number of poor people in urban and rural areas in South Sulawesi Province, so an analysis is needed to classify areas that have the same characteristics based on poverty indicators. For this reason, these three methods are used. However, the results of this analysis are only based on the similarity measure based on the distance technique used. Thus, the multiscale bootstrap method is used to obtain the validity of the resulting clusters. The results of the research using these three methods are four clusters with different characteristics. By using multiscale bootstrap, it is found that in single linkage there are four valid clusters, for complete linkage there is only one valid cluster and on average linkage there are three valid clusters. So it is found that single linkage is the best method in classifying these cases.
Regresi Data Panel dan Aplikasinya dalam Kinerja Keuangan terhadap Pertumbuhan Laba Perusahaan Idx Lq45 Bursa Efek Indonesia Nurul Madany; Ruliana Ruliana; Zulkifli Rais
VARIANSI: Journal of Statistics and Its application on Teaching and Research Vol. 4 No. 2 (2022)
Publisher : Program Studi Statistika Fakultas MIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (391.985 KB) | DOI: 10.35580/variansiunm28

Abstract

Regression is a statistical analysis that shows the relationship between one bound changer and one or more free changers. In the development of regression analysis, which can not only be observed at one time but can be observed in several time periods known as panel data regression. In conducting the regression analysis of panel data, there are three tests carried out to select a fixed model, namely the chow test, the Hausman test, and the pagan breucsh test. This study aims to see the influence of free variables, namely roa, roe, and npm on bound variables, namely company profit growth. The implementation of the method is carried out in the case of data on the financial performance of the LQ45 company and the growth of the company's profit LQ45. The result of the panel data regression modeling, namely the fixed effect model, is that financial performance has a significant effect on the company's profit growth whereas in the financial performance indicators the roa variable has a positive and significant influence and has a presentation that explains the free variables, namely roa, roe, and npm, on profit growth of the remaining 21% explained by other variables.
PENERAPAN METODE RANDOM FOREST UNTUK KLASIFIKASI VARIAN MINUMAN KOPI DI KEDAI KOPI KONIJIWA BANTAENG Suci Amaliah; Muhammad Nusrang; Aswi Aswi
VARIANSI: Journal of Statistics and Its application on Teaching and Research Vol. 4 No. 3 (2022)
Publisher : Program Studi Statistika Fakultas MIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (595.241 KB) | DOI: 10.35580/variansiunm31

Abstract

Random Forest (RF) adalah metode yang dapat meningkatkan hasil akurasi dalam membangkitkan atribut untuk setiap node yang dilakukan secara acak. Penelitian ini bertujuan untuk mengetahui tingkat akurasi metode RF dalam memprediksi varian minuman kopi di kedai Konijiwa Bantaeng yang paling diminati pelanggan. Berdasarkan hasil analisis diperoleh bahwa model dengan error klasifikasi terkecil adalah dengan menggunakan mtry 2 dan ntree 500. Model yang dihasilkan dievaluasi dengan menggunakan confusion matrix dimana diperoleh bahwa varian minuman kategori coffee based lebih diminati daripada signature coffee dengan nilai akurasi sebesar 94,12%.

Page 1 of 2 | Total Record : 16