Garuda - Garba Rujukan Digital

Aplikasi Generalized Poisson Regression dalam Mengatasi Overdispersi pada Data Jumlah Penderita Demam Berdarah Dengue Arwini Arisandi; Erna Tri Herdiani; Sitti Sahriman
STATISTIKA: Forum Teori dan Aplikasi Statistika Vol 18, No 2 (2018)
Publisher : Program Studi Statistika Unisba

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29313/jstat.v18i2.4542

Asumsi dasar dalam regresi Poisson yaitu nilai variansi data sama dengan nilai mean data. Namun,asumsi tersebut umumnya tidak terpenuhi, misalnya terdapat kasus overdispersi. Overdispersidalam regresi Poisson terjadi apabila nilai variansinya lebih besar daripada nilai meannya. Jikaterjadi overdispersi pada data, maka model regresi Poisson kurang akurat digunakan karenaberdampak pada nilai standard error dari taksiran parameter yang dihasilkan cenderung menjadiunderestimate sehingga kesimpulan yang diperoleh menjadi kurang valid. Dalam penelitian ini,kasus overdispersi dapat diatasi dengan model generalized Poisson regression. Hasil penelitianmenunjukkan bahwa nilai AIC minimum diberikan oleh model generalized Poisson regression.Sehingga dalam penelitian ini disimpulkan bahwa pada penelitian terhadap data yang mengalamioverdispersi pada Jumlah Penderita DBD di Kota Makassar tahun 2016, pemodelan regresigeneralized Poisson mampu mengatasi terjadinya overdispersi yang terjadi pada pemodelan regresiPoisson. Nilai R2 yang dimiliki sebesar 67% yang artinya jumlah penderita DBD ditentukan olehpersentase tempat-tempat umum memenuhi syarat kesehatan, persentase penduduk yang memilikiakses air minum layak, persentase rumah tangga berprilaku hidup bersih dan sehat dan persentaserumah yang memenuhi syarat kesehatan. Selebihnya 33% ditentukan oleh faktor lain.

Memprediksikan Indeks Pembangunan Manusia di Wilayah Indonesia Bagian Timur Menggunakan Random Forest Classification Arwini Arisandi; Syandriana Syarifuddin
Journal of Mathematics: Theory and Applications Volume 5, Nomor 1, 2023
Publisher : Program Studi Matematika Universitas Sulawesi Barat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31605/jomta.v5i1.2402

Abstrak. Indeks Pembangunan Manusia (IPM) merupakan salah satu indikator yang penting dalam melihat sisi lain dari pembangunan. Setiap indikator komponen penghitungan IPM dapat dimanfaatkan untuk mengukur keberhasilan pembangunan kualitas hidup manusia seperti Umur Harapan Hidup (UHH), Harapan Lama Sekolah (HLS), Pengeluaran per Kapita Disesuaikan (PKD), dan Lama Sekolah (LS). Penelitian ini bertujuan untuk mengetahui sebaran IPM di Kawasan Timur Indonesia, kemudian melakukan pemodelan data IPM dengan menggunakan regresi logistik, decision tree, dan random forest untuk mendapatkan model terbaik dalam memprediksi IPM serta mengetahui faktor-faktor yang memiliki pengaruh terhadap perubahan nilai IPM. Hasilnya menunjukkan bahwa daerah dengan kategori IPM rendah dan IPM sedang memiliki persentase sebesar 69% yang lebih tinggi dibandingkan dengan daerah dengan kategori IPM tinggi dan IPM sangat tinggi sebesar 31% untuk kawasan Timur Indonesia. Model terbaik untuk pemodelan data IPM pada Kawasan Timur Indonesia adalah model random forest dengan nilai kebaikan model sebesar 94.03% dan nilai balanced accuracy sebesar 93.33%. Hasil prediksi diperoleh sebanyak 2 kabupaten/kota atau 4.08% yang diprediksi tidak tepat. Variabel Umur Harapan Hidup memiliki pengaruh atau kontribusi yang signifikan dalam perubahan nilai IPM kabupaten/kota di Kawasan Timur Indonesia. Kata kunci: IPM, Kawasan Timur Indonesia, Random forest

PERBANDINGAN AKURASI MODEL FUZZY TIME SERIES DALAM PERAMALAN HARGA CABAI RAWIT MERAH DI KOTA MAKASSAR Gaffar, Ismail; Arisandi, Arwini; Makkulawu, A Ridwan
Buletin Sistem Informasi dan Teknologi Islam (BUSITI) Vol 5, No 2 (2024)
Publisher : Universitas Muslim Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33096/busiti.v5i2.2278

Cabai rawit (Capsicum frutescens) merupakan bagian dari famili solanaceae. Cabai rawit adalah salah satu tanaman hortikultura yang banyak dibudidayakan di Indonesia. Selain dikonsumsi, zat capsaicin pada cabai rawit juga digunakan sebagai bahan kosmetik. Harga cabai rawit terkadang meresahkan masyarakat dikarenakan lonjakan harga. Sehingga peramalan harga diperlukan pemerintah untuk mengatasi lonjakan harga yang terjadi setiap tahunnya. Peramalan (Forecasting) merupakan ilmu yang mampu mempelajari informasi historis yang dapat meramalkan kejadian yang akan datang. Ilmu peramalan menggunakan data runtun waktu (Time Series) untuk mempelajari data historis, dapat berupa data harian, bulan, ataupun periode waktu tertentu. Dalam penelitian ini, membandingkan Fuzzy Time Series (FTS) model Chen dan Lee untuk mengetahui performa yang terbaik. Data yang digunakan yaitu panel harga cabai rawit merah di Kota Makassar sejak Juli 2022 hingga Mei 2024. Masing-masing model memiliki cara tersendiri dalam membentuk Fuzzy Logic Relationship Group (FLRG). Hasil penelitian menunjukkan model Lee memberikan performa terbaik untuk meramalkan harga cabai rawit merah dengan nilai Mean Absolute Percentage Error (MAPE) sebesar 6,00%.

Optimization of Drying Models for Various Types of Turmeric Using A Tray Dryer Jassin, Ernawati; Fitri, Muhammad; Arisandi, Arwini; Aisyah, Nur
Journal of Agriculture Vol. 3 No. 03 (2024): Research Articles November 2024
Publisher : ITScience (Information Technology and Science)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/joa.v3i03.4911

Turmeric (Curcuma domestica val) is one of the plants that has benefits including as spices, seasonings, herbs, to maintain health and beauty and utilization as traditional medicine. Drying is a method to extend the shelf life and maintain the quality of a post-harvest product before further processing. This study aims to optimize the thin layer drying model on various types of turmeric (white turmeric, black turmeric and yellow turmeric). Data obtained from drying results in the form of initial mass, mass during drying and the final mass of drying. Then processed to obtain the wet base moisture content, dry base moisture content and Moisture Ratio. The thin layer drying model is obtained by finding the constant value of k, a and n from each exponential form. Determination of constants using Solver Tool Microsoft Excel that automatically searches for constant values in each dryer model tested, then obtains the highest R² value as the best model that describes the drying model of black turmeric and yellow turmeric. From the test results of curcumin content in white turmeric, black turmeric and yellow turmeric where the content of yellow turmeric has a higher value than both turmeric because it has an almost perfect arrangement of chemical elements. In the pH test, the highest result was obtained in white turmeric, namely 4.8 where the less curcumin contained in the turmeric rhizome, the higher the level of acidity of the turmeric.

Optimization of Drying Models for Various Types of Turmeric Using A Tray Dryer Jassin, Ernawati; Fitri, Muhammad; Arisandi, Arwini; Aisyah, Nur
Journal of Agriculture Vol. 3 No. 03 (2024): Research Articles November 2024
Publisher : ITScience (Information Technology and Science)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/joa.v3i03.4911

Turmeric (Curcuma domestica val) is one of the plants that has benefits including as spices, seasonings, herbs, to maintain health and beauty and utilization as traditional medicine. Drying is a method to extend the shelf life and maintain the quality of a post-harvest product before further processing. This study aims to optimize the thin layer drying model on various types of turmeric (white turmeric, black turmeric and yellow turmeric). Data obtained from drying results in the form of initial mass, mass during drying and the final mass of drying. Then processed to obtain the wet base moisture content, dry base moisture content and Moisture Ratio. The thin layer drying model is obtained by finding the constant value of k, a and n from each exponential form. Determination of constants using Solver Tool Microsoft Excel that automatically searches for constant values in each dryer model tested, then obtains the highest R² value as the best model that describes the drying model of black turmeric and yellow turmeric. From the test results of curcumin content in white turmeric, black turmeric and yellow turmeric where the content of yellow turmeric has a higher value than both turmeric because it has an almost perfect arrangement of chemical elements. In the pH test, the highest result was obtained in white turmeric, namely 4.8 where the less curcumin contained in the turmeric rhizome, the higher the level of acidity of the turmeric.

APPLICATION OF THE ARIMA METHOD IN FORECASTING THE PRICE RED CAYENNE PEPPER IN MAKASSAR CITY Arisandi, Arwini; Gaffar, Ismail; Makkulawu, Andi Ridwan
Parameter: Journal of Statistics Vol. 4 No. 1 (2024)
Publisher : Fakultas Matematika dan Ilmu Pengetahuan Alam Universitas Tadulako

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22487/27765660.2024.v4.i1.17120

Red chili is one of the commodities with very tall cost changes. The cost variance of red chili can be caused by a huge amount of supply and request. The higher the amount of supply, the lower the cost, and the lower the amount of supply, the higher the cost. This study aims to implement the ARIMA method in forecasting red cayenne pepper prices in Makassar City. Data analysis to forecast red cayenne pepper prices used the ARIMA method with the results show that the price range of chili is from IDR 13,000 to IDR 80,000, with a mean value of IDR 38,218. The model with the minimum SSE and MSE value is ARIMA(1,1,1), so this model be used in time series data modeling for forecasting. The results of forecasting using the best model obtained a MAPE value of 15.90%, which is in the range of 10-20%, so it can be concluded that the ability of the ARIMA(1,1,1) model in forecasting the price of red cayenne pepper includes the good category.

Implementation of Random Forest Algorithm for Shallot Price Forecasting in Makassar City Hardianti Hafid; Arwini Arisandi; Reski Wahyu Yanti
Journal of Mathematics, Computations and Statistics Vol. 8 No. 2 (2025): Volume 08 Nomor 02 (Oktober 2025)
Publisher : Jurusan Matematika FMIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35580/jmathcos.v8i2.9477

This study aims to implement the Random Forest algorithm for forecasting shallot prices in Makassar City using monthly historical data from January 2018 to December 2024, obtained from the Statistics Indonesia (Badan Pusat Statistik) of South Sulawesi Province. The analysis begins with identifying significant lags through the Partial Autocorrelation Function (PACF) plot, resulting in seven input variable schemes. Each scheme was tested using training and testing datasets. Model performance was evaluated using the Mean Absolute Percentage Error (MAPE). The results show that Scheme 1 (Lag 1) achieved the best performance with a MAPE value of 13.08%, which falls into the “good” category. Price forecasts for January–December 2025 using the best scheme indicate a price range of IDR 23,200 – 24,300 per kilogram, with peak prices in March, July, and November, and the lowest prices in April, August, and December. Although the model successfully captures historical price patterns, real-world fluctuations driven by seasonal factors, supply disruptions, and distribution costs may cause prediction deviations. This study recommends integrating exogenous variables and real-time data to improve forecasting accuracy and support local food price stabilization policies.

Effect of Sugar and Juice Concentration on Vitamin C Retention and Organoleptic of Tangerine (Citrus Reticulata) Powder Ahmad, Ilham; Jassin, Ernawati; Arisandi, Arwini; Mus, Rahmi; Makkulawu, Andi Ridwan
Journal of Agriculture Vol. 4 No. 03 (2025): Research Articles, November 2025
Publisher : ITScience (Information Technology and Science)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/joa.v4i03.6951

Converting tangerine juice into powder via sucrose crystallization yields shelf?stable products with appealing flavor and nutrients. We examined the independent and interactive effects of sucrose and juice concentrations on physicochemical properties and consumer acceptance of Citrus reticulata powder. A 3×3 factorial design evaluated sucrose at 45, 55, and 65% and juice at 45, 55, and 65% (three replications). Processing included extraction and filtration, sucrose addition, controlled heating to supersaturation, crystallization during cooling, milling, and sieving. Moisture content, vitamin C (UV–Vis, 245 nm), and hedonic ratings for texture, aroma, color, and taste (nine?point scale; 30 panelists) were measured. ANOVA and Duncan’s Multiple Range Test assessed treatment effects. Moisture was 2.04–3.07% across treatments and did not differ significantly (p=0.924), meeting the SNI specification for powdered beverages (<5%). Vitamin C differed significantly (p=0.006); the 45% sucrose level consistently produced the highest values (?42.9–43.8 mg/100 g). The A1B3 formulation (45% sucrose, 65% juice) combined high vitamin C (~43.2 mg/100 g) with acceptable moisture (3.07%) and the most preferred sensory profile. Patterns suggest that reducing sucrose limits thermal/Maillard losses of ascorbic acid, while very high sucrose lowers water activity and yields intermediate retention. Overall, sucrose crystallization produced shelf?stable powders with favorable reconstitution, nutritional, and liking properties. A low?sucrose, high?juice formula optimizes the balance of quality attributes, supporting use in instant citrus drinks and functional beverages.

Akurasi Model Prediksi Menggunakan Metode Automatic Clustering Fuzzy Time Series pada Indeks Harga Konsumen di Kota Makassar Arisandi, Arwini; Hafid, Hardianti
Journal of Mathematics: Theory and Applications Vol 6 No 1 (2024): Volume 6, Nomor 1, 2024
Publisher : Program Studi Matematika Universitas Sulawesi Barat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31605/jomta.v6i1.3621

Indeks Harga Konsumen (IHK) adalah sebagai alat pengukuran yang digunakan untuk memantau perubahan harga barang dan jasa yang dibeli oleh rumah tangga konsumen dalam suatu periode waktu tertentu. Informasi IHK ini merupakan alat ukur yang digunakan oleh Badan Pusat Statistik (BPS) untuk mengetahui nilai inflasi pada suatu periode tertentu sehingga memprediksikan IHK dapat mengontrol laju inflasi di suatu daerah. Oleh karena itu, penelitian ini bertujuan untuk mengetahui akurasi model prediksi menggunakan metode Automatic Clustering Fuzzy Time Series pada IHK di Kota Makassar. Akurasi model prediksi diukur melalui nilai mean square error (MSE) dan mean absolute percentage error (MAPE). Data sekunder diperoleh dari BPS dengan rentang waktu Januari 2020 hingga November 2023. Hasilnya menunjukkan bahw nilai MSE yang diperoleh adalah 0,059 dan nilai MAPE sebesar 0,154%. Hal ini menunjukkan bahwa nilai MAPE berada pada rentang <10% yang disimpulkan bahwa kemampuan metode Automatic Clustering Fuzzy Time Series sangat baik dalam memprediksikan IHK di Kota Makassar.

Klasifikasi Penggunaan Teknologi Pada Petani Milenial di Sulawesi Selatan Menggunakan Density Based Spatial Clustering Algorithm With Noise Hafid, Hardianti; Arisandi, Arwini
Journal of Mathematics: Theory and Applications Vol 6 No 1 (2024): Volume 6, Nomor 1, 2024
Publisher : Program Studi Matematika Universitas Sulawesi Barat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31605/jomta.v6i1.3623

This research examines the use of agricultural technology by millennial farmers in South Sulawesi Province by applying the Density-Based Spatial Clustering Algorithm with Noise (DBSCAN) to the agricultural census data of South Sulawesi Province Phase 1 in 2023. The secondary data used includes the number of millennial farmers aged 19-39 who use or do not use digital technology, divided by district/city and gender. The analysis process begins with preprocessing to prepare the data, followed by clustering using the DBSCAN algorithm, determining the optimal values for the Eps and minPts parameters, and evaluating the quality of the formed clusters using the silhouette coefficient and elbow method. The results of the study indicate that the combination of Eps value of 1.000 and minPts value of 7 produces optimal clustering with 2 clusters formed and 92 data points clustered, while 4 other data points are considered as noise. Evaluation using the silhouette coefficient and elbow method also indicates that the optimal data grouping is k=2.

Title

Found 13 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 13 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 13 Documents
Search