Muhammad Fahmuddin S
Department of Statistics, Universitas Negeri Makassar

Published : 5 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 5 Documents
Search

Analisis Regresi Nonparametrik Spline Truncated untuk Menganalisis Faktor-Faktor yang Mempengaruhi Tingkat Pengangguran Terbuka di Provinsi Sulawesi Selatan Devi Carolin Wongkar; Ruliana Ruliana; Muhammad Fahmuddin S
VARIANSI: Journal of Statistics and Its application on Teaching and Research Vol. 5 No. 02 (2023)
Publisher : Program Studi Statistika Fakultas MIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35580/variansiunm101

Abstract

The nonparametric regression analysis is a regression model used to determine the relationship between response variable and independent variables with unknown regression curve shapes. In the nonparametric approach, one of the frequently used estimators is the spline truncated. Spline truncated model is a segmented polynomial truncation model. The advantage of this model is that it is flexible because it has knot points that can show changes in data patterns. The unemployment rate in South Sulawesi Province in 2021 reached 5.72% and became the province with the second highest unemployment rate on Sulawesi Island. Therefore, spline truncated nonparametric regression modelling will be carried out in the case of unemployment rate with each of the factors that are thought to be influential because the regression curve is found not to form a certain pattern. Based on the analysis results, the best truncated spline nonparametric regression model was obtained using three knot points and obtained the minimum GCV value of 0.38 with a coefficient of determination (R2) value of 89%. Factors that have a significant effect on the unemployment rate in South Sulawesi are mean years of schooling (x1) and labour force participation rate (x2).
APLIKASI FUNGSI TRANSFER MULTIVARIAT UNTUK PERAMALAN CURAH HUJAN DI KOTA MAKASSAR Idul Fitri Abdullah Abdullah; Ruliana Ruliana; Muhammad Fahmuddin S
VARIANSI: Journal of Statistics and Its application on Teaching and Research Vol. 5 No. 02 (2023)
Publisher : Program Studi Statistika Fakultas MIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35580/variansiunm117

Abstract

This study aims to determine the transfer function model and factors that significantly affect the level of rainfall in Makassar city. This study uses rainfall data as the output series and air humidity (X1), air temperature (X2) and wind speed (X3) as the input series. The data used is monthly data with a period of January 2013 - December 2022. The initial stage of modeling is done by determining the ARIMA model of each input series which is then used to calculate the identification of the transfer function model Based on the research obtained multivariate transfer function model X1 (b=3, r=0, s=0) X12(b=0, r=0, s=0) ARIMA (2,1,0)(1,1,0)12 with air humidity and air temperature being significant factors affecting rainfall in Makassar city.
Analisis Model Volatilitas ARCH/GARCH untuk Mengestimasi Value at Risk pada Data Saham PT Indofood CBP Sukses Makmur Tbk (ICBP) Muhammad Nusrang; Muhammad Fahmuddin S; Dhiya Izdihar Haryadi
VARIANSI: Journal of Statistics and Its application on Teaching and Research Vol. 7 No. 03 (2025)
Publisher : Program Studi Statistika Fakultas MIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35580/variansiunm413

Abstract

Dalam data time series keuangan, volatilitas cenderung tinggi dan ditandai dengan fluktuasi besar, yaitu periode volatilitas tinggi diikuti periode rendah, kemudian kembali meningkat, atau dikenal sebagai masalah heteroskedastisitas. Pemodelan ARIMA Box-Jenkins tidak lagi valid sehingga diperlukan metode alternatif seperti ARCH/GARCH. Penelitian ini bertujuan memodelkan volatilitas harga penutupan saham PT ICBP menggunakan ARCH/GARCH serta mengestimasi nilai risiko investasi dengan Value at Risk (VaR). Hasil analisis menunjukkan bahwa model ARIMA (0,1,1)-ARCH (1) merupakan model terbaik dalam menggambarkan volatilitas saham dengan parameter signifikan dan akurasi peramalan yang cukup baik (MAPE = 5,8%). Estimasi VaR dengan simulasi historis menunjukkan potensi kerugian maksimum harian untuk investasi Rp10.000.000,00 sebesar Rp534,48 pada tingkat keyakinan 90%, Rp541,17 pada 95%, dan Rp546,54 pada 99%. Temuan ini memberikan gambaran jelas mengenai risiko harian yang dapat dijadikan bahan pertimbangan dalam pengambilan keputusan investasi pada saham ICBP.
PENERAPAN METODE HYBRID SSA-ARIMA PADA PERAMALAN INDEKS HARGA KEBUTUHAN PERTANIAN YANG DIBAYAR PETANI DI PROVINSI SULAWESI SELATAN: indonesia Muhammad Fahmuddin S; Ruliana; Muh. Imam Shadiq
VARIANSI: Journal of Statistics and Its application on Teaching and Research Vol. 7 No. 02 (2025)
Publisher : Program Studi Statistika Fakultas MIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35580/variansiunm317

Abstract

This study aims to determine the results and accuracy of forecasting the farmer's price index (IHDP) in South Sulawesi Province using Hybrid SSA-ARIMA. Hybrid SSA-ARIMA is a combination of two good time series methods to improve forecasting accuracy, especially for IHDP data that contains trend and seasonal elements. The data used is the South Sulawesi IHDP data from January 2019 to June 2024 which is sourced from the official website of the Central Statistics Agency. The results of the IHDP forecast in South Sulawesi for the next 12 months from July 2023 to June 2024 tend to increase with the largest increase in September 2024 of 1.184 with a forecast accuracy based on the Mean Absolute Percentage Error (MAPE) of 1.59%. This shows that Hybrid SSA-ARIMA has very good forecasting capabilities
APPLICATION OF TIME SERIES REGRESSION (TSR) AND AUTOREGRESSIVE INTEGRATED MOVING AVERAGE (ARIMA) IN RICE PRODUCTION FORECASTING IN INDONESIA Muhammad Fahmuddin S; Ruliana; Nurul Fahmi
VARIANSI: Journal of Statistics and Its application on Teaching and Research Vol. 7 No. 03 (2025)
Publisher : Program Studi Statistika Fakultas MIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35580/variansiunm412

Abstract

Rice production plays a crucial role in supporting food security in Indonesia. The annual fluctuations in rice yield necessitate accurate forecasting methods to support agricultural planning. This study aims to forecast rice production in Indonesia using two time series forecasting approaches: Time Series Regression (TSR) and Autoregressive Integrated Moving Average (ARIMA). The data used consist of monthly rice production from January 2020 to December 2024. The analysis results show that both methods are capable of modeling the data well, with high forecasting accuracy based on the Mean Absolute Percentage Error (MAPE). The TSR model yielded a MAPE of 13.838%, while the ARIMA(2,1,0)(0,1,0)12model achieved a lower MAPE of 13.1439%, indicating that the ARIMA model provides more accurate forecasting results. This study is expected to serve as a reference for policy-making and strategic planning in rice production management in the future.