Clustering of Banking Sector Stocks using Integration of Fourier Transform, Spectral Clustering, and Fuzzy C-Means as a Basis for Mean-Variance Portfolio Optimization

Dimitri Salsabila Fakhriyah Ricardo; Nurul Gusriani; Herlina Napitupulu

doi:10.24198/jmi.v22.n1.69287.59-72

Clustering of Banking Sector Stocks using Integration of Fourier Transform, Spectral Clustering, and Fuzzy C-Means as a Basis for Mean-Variance Portfolio Optimization

Dimitri Salsabila Fakhriyah Ricardo, Nurul Gusriani, Herlina Napitupulu

Abstract

The Indonesian capital market has experienced significant growth accompanied by high volatility, particularly in the banking sector which
holds a substantial contribution to market capitalization. Extreme
volatility during the 2019-2024 period triggered by the impact of the
COVID-19 pandemic, economic recovery phases, and global macroeconomic challenges has created complexities in investment decisionmaking and portfolio optimization, which often produces unstable solutions under uncertain market conditions. Various studies have applied frequency domain analysis to uncover hidden patterns in stock
price movements or clustering to group stocks based on their characteristics to support optimal investment decision-making, however the
integration of these two approaches remains limited in its application to
generate robust portfolio optimization solutions in the Indonesian capital market. This study aims to generate clustering of banking sector
stocks through the integration of Fourier Transform, spectral clustering, and Fuzzy C-Means and to construct an optimal portfolio using the
Mean-Variance method based on the clustering results. This study uses
closing price data of 41 banking sector stocks on the Indonesia Stock
Exchange for the 2019-2024 period through an integrated approach of
Fourier Transform to extract frequency patterns, spectral clustering as
a basis for grouping, Fuzzy C-Means to generate cluster membership degrees, and Mean-Variance for portfolio optimization. The results show
that the integration of these methods produces four optimal stock clusters consisting of nine stocks with a medium risk-low return profile,
six stocks with a high risk-high return profile, fifteen stocks with a low
risk-low return profile, and eleven stocks with a medium risk-high return profile. Based on the clustering results, four representative stocks
were selected from each cluster for portfolio optimization, resulting in
an optimal portfolio at a risk aversion value of ρ = 6.83 with a portfolio ratio of 3.4128877. This optimal portfolio is constructed from four
representative stocks with weight allocations of 11.48% BMAS, 11.76%
ARTO, 72.08% BNGA, and 4.68% BBHI, with an expected return value
of 0.0263613 and a portfolio variance of 0.0077241.

Keywords

Banking Stocks, Clustering, Fourier Transform, Spectral Clustering, Fuzzy C-Means, Mean-Variance

Full Text:

PDF

References

Kusuma, A.A., Farhan, M., Maedianda, M., and Rahmah, F., 2024, Kebijakan moneter terhadap stabilitas pasar modal di Indonesia, EQUILIBRIUM: Jurnal Ilmiah Ekonomi dan Pembelajarannya, 12(1), pp. 55–64. doi: 10.25273/equilibrium.v12i1.18701.

Bursa Efek Indonesia, 2019, Laporan Tahunan 2019, Bursa Efek Indonesia, https://www.idx.co.id/id/tentang-bei/laporan-tahunan/.

Bursa Efek Indonesia, 2024, Laporan Tahunan 2024, Bursa Efek Indonesia, https://www.idx.co.id/id/tentang-bei/laporan-tahunan/.

Bursa Efek Indonesia, 2023, Tambah Alternatif Acuan Investasi Sub-Sektor Bank, BEI dan PEFINDO Luncurkan Indeks IDX-PEFINDO Prime Bank, Bursa Efek Indonesia, https://www.idx.co.id/id/berita/siaran-pers/2028.

Lyle, M.R., and Yohn, T.L., 2020, Fundamental analysis and Mean-Variance optimal portfolios, The Accounting Review, 96(6), pp. 303–327. doi: 10.2308/TAR-2019-0622.

Scalzo, B., Arroyo, A., Stanković, L., and Mandic, D.P., 2021, Nonstationary portfolios: Diversification in the spectral domain, ICASSP 2021 – IEEE International Conference on Acoustics, Speech and Signal Processing, pp. 5155–5159. doi: 10.1109/ICASSP39728.2021.9413769.

Alam, K.E., Saputra, K.V.I., and Margaretha, H., 2021, Stocks clustering with Fourier transformation towards Indonesian stocks for pairs trading and investment diversification, AIP Conference Proceedings, 020033. doi: 10.1063/5.0075346.

Gubu, L., Cahyono, E., Arman, A., Budiman, H., and Djafar, M.K., 2023, Optimasi portofolio Mean-Variance dengan analisis klaster Fuzzy C-Means, Jurnal Gaussian, 12(4), pp. 593–604. doi: 10.14710/j.gauss.12.4.593-604.

Sunariyah, S., 2006, Pengantar pengetahuan pasar modal, 5th ed., Yogyakarta: UPP STIM YKPN.

Anoraga, P. and Pakarti, P., 2008, Pengantar pasar modal, Jakarta: Rineka Cipta.

Elawati, E., Hayati, R., and Hanafi, H., 2020, Analisis perbandingan metode Descrete Fourier Transform dan metode Descrete Cosine Transform pada teknik menyembunyikan sinyal suara, Jurnal Tektro, 4(2), pp. 82–90. doi: 10.30811/tektro.v6i1.3219.

Yusuf, A., and Tjandrasa, H., 2014, Prediksi nilai dengan metode spectral clustering dan clusterwise regression, Jurnal Simantec, 4(1), pp. 1–8.

Sharma, A., Kumar, R., and Mansotra, V., 2016, Proposed stemming algorithm for Hindi information retrieval, International Journal of Innovative Research in Computer and Communication Engineering, 3297(6), pp. 11449–11455. doi: 10.15680/IJIRCCE.2016.

von Luxburg, U., 2007, A tutorial on spectral clustering, Statistics and Computing, 17(4), pp. 395–416. doi: 10.1007/s11222-007-9033-z.

Wulandari, S., 2020, Clustering microarray adenoma menggunakan spectral clustering dengan algoritma Partitioning Around Medoid (PAM), Prosiding Seminar Nasional Sains 2020, pp. 345–351.

Aliansa, W., and Saputra, R.A., 2023, Perbandingan harga saham perbankan dengan metode Fuzzy C-Means clustering, Prosiding Seminar Nasional Pemanfaatan Sains dan Teknologi Informasi, pp. 369–376.

Rahmatika, L., Suparti, S., and Safitri, D., 2015, Analisis kelompok dengan algoritma Fuzzy C-Means dan Gustafson Kessel clustering pada Indeks LQ45, Jurnal Gaussian, 4(3), pp. 543–551.

Bezdek, J.C., 1981, Pattern recognition with fuzzy objective function algorithms, New York: Plenum Press.

Ningrat, D.R., Maruddani, D.A.I., and Wuryandari, T., 2016, Analisis cluster dengan algoritma K-Means dan Fuzzy C-Means clustering untuk pengelompokan data obligasi korporasi, Jurnal Gaussian, 5(4), pp. 641–650. doi: 10.14710/j.gauss.5.4.641-650.

Guo, Q., 2022, Review of research on Markowitz model in portfolios, 2022 7th International Conference on Social Sciences and Economic Development (ICSSED 2022), pp. 786–790. doi: 10.2991/aebmr.k.220405.131.

DOI: https://doi.org/10.24198/jmi.v22.n1.69287.59-72