Javascript must be enabled to continue!
Penerapan Metode Hybrid Fuzzy Time Series pada Data IHSG
View through CrossRef
Abstract. Forecasting the Indonesia Composite Stock Price Index (IHSG) plays a crucial role in investment decision-making; however, the volatility and uncertainty of the data make the process complex. This study aims to apply a hybrid Fuzzy Time Series (FTS) method combined with Fuzzy C-Means (FCM) and Markov Chain algorithms to accurately model and forecast IHSG values. The data used consists of daily IHSG closing prices from January 2020 to July 2025, obtained from idx.co.id. The research steps include data preprocessing, determining the universe of discourse, identifying the optimal number of clusters using the Elbow method, constructing fuzzy intervals based on FCM cluster centers, followed by fuzzification, forming Fuzzy Logical Relationship Groups (FLRG), building the Markov transition probability matrix, and conducting prediction with result adjustments. The model was evaluated using RMSE, MAPE, and Theil’s U metrics to measure accuracy. The results showed that the hybrid method produced predictions with an RMSE of 65.8147, MAPE of 0.80%, and Theil’s U of 0.0102 after adjustment, indicating excellent model performance. In conclusion, the FTS-FCM-Markov hybrid method effectively handles the dynamic characteristics of IHSG data and can serve as an alternative approach for stock market forecasting.
Abstrak. Peramalan Indeks Harga Saham Gabungan (IHSG) memiliki peran penting dalam pengambilan keputusan investasi, namun fluktuasi dan ketidakpastian data membuat proses ini menjadi kompleks. Penelitian ini bertujuan untuk menerapkan metode hybrid Fuzzy Time Series (FTS) yang dikombinasikan dengan algoritma Fuzzy C-Means (FCM) dan Markov Chain untuk memodelkan dan meramalkan nilai IHSG secara akurat. Data yang digunakan merupakan harga penutupan IHSG harian dari Januari 2020 hingga Juli 2025 yang diambil dari Idx.co.id. Langkah penelitian dimulai dengan pra-pemrosesan data, penentuan universe of discourse, penentuan jumlah klaster optimal dengan metode Elbow, pembentukan interval fuzzy berdasarkan pusat klaster FCM, kemudian dilakukan proses fuzzifikasi, pembentukan Fuzzy Logical Relationship Group (FLRG), pembangunan matriks probabilitas transisi Markov, hingga tahap prediksi dan penyesuaian hasil. Model dievaluasi menggunakan metrik RMSE, MAPE, dan Theil’s U untuk mengukur tingkat akurasi. Hasil menunjukkan bahwa metode hybrid ini mampu memberikan prediksi dengan RMSE sebesar 65.8147, MAPE 0.80%, dan Theil’s U sebesar 0.0102 setelah dilakukan penyesuaian, yang mengindikasikan performa model yang sangat baik. Kesimpulannya, metode hybrid FTS-FCM-Markov terbukti mampu menangani karakteristik data IHSG yang bersifat dinamis dan dapat dijadikan sebagai pendekatan alternatif dalam peramalan pasar saham..
Universitas Islam Bandung (Unisba)
Title: Penerapan Metode Hybrid Fuzzy Time Series pada Data IHSG
Description:
Abstract.
Forecasting the Indonesia Composite Stock Price Index (IHSG) plays a crucial role in investment decision-making; however, the volatility and uncertainty of the data make the process complex.
This study aims to apply a hybrid Fuzzy Time Series (FTS) method combined with Fuzzy C-Means (FCM) and Markov Chain algorithms to accurately model and forecast IHSG values.
The data used consists of daily IHSG closing prices from January 2020 to July 2025, obtained from idx.
co.
id.
The research steps include data preprocessing, determining the universe of discourse, identifying the optimal number of clusters using the Elbow method, constructing fuzzy intervals based on FCM cluster centers, followed by fuzzification, forming Fuzzy Logical Relationship Groups (FLRG), building the Markov transition probability matrix, and conducting prediction with result adjustments.
The model was evaluated using RMSE, MAPE, and Theil’s U metrics to measure accuracy.
The results showed that the hybrid method produced predictions with an RMSE of 65.
8147, MAPE of 0.
80%, and Theil’s U of 0.
0102 after adjustment, indicating excellent model performance.
In conclusion, the FTS-FCM-Markov hybrid method effectively handles the dynamic characteristics of IHSG data and can serve as an alternative approach for stock market forecasting.
Abstrak.
Peramalan Indeks Harga Saham Gabungan (IHSG) memiliki peran penting dalam pengambilan keputusan investasi, namun fluktuasi dan ketidakpastian data membuat proses ini menjadi kompleks.
Penelitian ini bertujuan untuk menerapkan metode hybrid Fuzzy Time Series (FTS) yang dikombinasikan dengan algoritma Fuzzy C-Means (FCM) dan Markov Chain untuk memodelkan dan meramalkan nilai IHSG secara akurat.
Data yang digunakan merupakan harga penutupan IHSG harian dari Januari 2020 hingga Juli 2025 yang diambil dari Idx.
co.
id.
Langkah penelitian dimulai dengan pra-pemrosesan data, penentuan universe of discourse, penentuan jumlah klaster optimal dengan metode Elbow, pembentukan interval fuzzy berdasarkan pusat klaster FCM, kemudian dilakukan proses fuzzifikasi, pembentukan Fuzzy Logical Relationship Group (FLRG), pembangunan matriks probabilitas transisi Markov, hingga tahap prediksi dan penyesuaian hasil.
Model dievaluasi menggunakan metrik RMSE, MAPE, dan Theil’s U untuk mengukur tingkat akurasi.
Hasil menunjukkan bahwa metode hybrid ini mampu memberikan prediksi dengan RMSE sebesar 65.
8147, MAPE 0.
80%, dan Theil’s U sebesar 0.
0102 setelah dilakukan penyesuaian, yang mengindikasikan performa model yang sangat baik.
Kesimpulannya, metode hybrid FTS-FCM-Markov terbukti mampu menangani karakteristik data IHSG yang bersifat dinamis dan dapat dijadikan sebagai pendekatan alternatif dalam peramalan pasar saham.
Related Results
Konstruksi Sistem Inferensi Fuzzy Menggunakan Subtractive Fuzzy C-Means pada Data Parkinson
Konstruksi Sistem Inferensi Fuzzy Menggunakan Subtractive Fuzzy C-Means pada Data Parkinson
Abstract. Fuzzy Inference System requires several stages to get the output, 1) formation of fuzzy sets, 2) formation of rules, 3) application of implication functions, 4) compositi...
Generated Fuzzy Quasi-ideals in Ternary Semigroups
Generated Fuzzy Quasi-ideals in Ternary Semigroups
Here in this paper, we provide characterizations of fuzzy quasi-ideal in terms of level and strong level subsets. Along with it, we provide expression for the generated fuzzy quasi...
New Approaches of Generalised Fuzzy Soft sets on fuzzy Codes and Its Properties on Decision-Makings
New Approaches of Generalised Fuzzy Soft sets on fuzzy Codes and Its Properties on Decision-Makings
Background Several scholars defined the concepts of fuzzy soft set theory and their application on decision-making problem. Based on this concept, researchers defined the generalis...
New Approaches of Generalised Fuzzy Soft sets on fuzzy Codes and Its Properties on Decision-Makings
New Approaches of Generalised Fuzzy Soft sets on fuzzy Codes and Its Properties on Decision-Makings
Background Several scholars defined the concepts of fuzzy soft set theory and their application on decision-making problem. Based on this concept, researchers defined the generalis...
Perbaikan Kualitas Citra Menggunakan Metode Fuzzy Type-2
Perbaikan Kualitas Citra Menggunakan Metode Fuzzy Type-2
Image enhancement is applied to an image that has low contrast. Histogram Equalization (HE) is a general method used to improve the quality of an image. However, its drawback is f...
DAMPAK TEKNOLOGI TERHADAP PROSES BELAJAR MENGAJAR
DAMPAK TEKNOLOGI TERHADAP PROSES BELAJAR MENGAJAR
DAFTAR PUSTAKAAditama, M. H. R., & Selfiardy, S. (2022). Kehidupan Mahasiswa Kuliah Sambil Bekerja di Masa Pandemi Covid-19. Kidspedia: Jurnal Pendidikan Anak Usia Dini, 3(...
ANALISIS CONTAGION EFFECT RESESI GLOBAL KE INDEKS PASAR MODAL ASEAN
ANALISIS CONTAGION EFFECT RESESI GLOBAL KE INDEKS PASAR MODAL ASEAN
The Covid-19 pandemic, the Russia-Ukraine war, and the increase in inflation in the United States have triggered a recession, leading to corrections in the global stock markets, in...
Peramalan Data IHSG Menggunakan Metode Backpropagation
Peramalan Data IHSG Menggunakan Metode Backpropagation
Jaringan saraf tiruan merupakan salah satu metode soft computing yang banyak digunakan dan diterapkan di berbagai disiplin ilmu, termasuk analisis data runtun waktu. Tujuan utama d...

