Kurniawan, Aznovri (2025) Predictive Analytics dengan Optimized Markov Switching Model dan Analisis Finansial untuk Strategi Pembiayaan Berkelanjutan bagi Perbankan Syariah di Indonesia. Doctoral thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 7032221017-Doctoral.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
7032221017-Doctoral.pdf - Accepted Version Restricted to Repository staff only Download (5MB) | Request a copy |
Abstract
Perbankan Syariah di Indonesia memiliki aset lebih kecil, non-performing financing (NPF) lebih tinggi, return on assets (ROA) lebih rendah, dan ketergantungan yang lebih tinggi terhadap pembiayaan dibandingkan total industri perbankan. Perbankan Syariah secara teori memiliki keterkaitan langsung dengan keuangan berkelanjutan (sustainable finance), karena adanya maqashid syariah yang menjadi pedoman, menunjukkan pentingnya penelitian ini terhadap isu global. Predictive dan prescriptive analytics telah digunakan dalam berbagai bidang, tetapi belum ditemukan referensi penggunaannya dalam penyusunan strategi pembiayaan perbankan. Berdasarkan data bulanan perbankan dari Otoritas Jasa Keuangan dan data keuangan dari Bank Indonesia tahun 2015-2023, penelitian ini mengembangkan model baru kombinasi linear programming (LP) untuk optimalisasi pendapatan, predictive analytics dengan Markov switching vector auto-regressive model (MS-VAR) disertai perbandingan dengan vector autoregressive (VAR) dan machine learning (ML), serta analisis finansial untuk simulasi pendapatan dan profitabilitas. Penelitian ini juga melakukan analisis empiris ketahanan perbankan Syariah terhadap krisis dan dukungan kepada sustainable finance yang dicanangkan secara global melalui Sustainable Development Goals (SDG). Hasil optimalisasi menunjukkan dukungan perbankan Syariah terhadap sustainable finance dan jika dilaksanakan dapat menghasilkan peningkatan ROA sebesar 32% di tahun 2024. Profitabilitas perbankan Syariah sejalan dengan agenda sustainability dan memiliki ketahanan yang lebih baik tetapi kurang antisipatif terhadap krisis daripada perbankan umum. Model LP menjadi basis untuk optimalisasi forecasting tahun 2024-2026 dengan analisis MS-VAR yang digabungkan dengan VAR atau ML. Strategi ini diproyeksikan akan menghasilkan peningkatan ROA sebesar rata-rata 25% per tahun selama 2024-2026, melebihi rata-rata pertumbuhan ROA total industri perbankan. Penelitian ini juga merekomendasikan perbankan Syariah untuk fokus kepada usaha mikro, kecil, dan menengah (UMKM) dan pembiayaan konsumsi, disertai beberapa fokus lapangan usaha pembiayaan, karena berkontribusi besar terhadap peningkatan profitabilitas. Penelitian ini menunjukkan bahwa model gabungan yang direkomendasikan sangat tepat untuk digunakan oleh perbankan Syariah. Melalui feedback mechanism dengan analisis finansial sehingga membentuk model prescriptive analytics, model ini menghasilkan strategi pembiayaan Syariah yang handal dan mampu mengakomodasi perubahan kondisi ekonomi. Penelitian lanjutan dan implementasi hasil penelitian ini dapat digunakan dengan menggunakan data untuk spesifik bank, yang biasanya berupa data harian dan berukuran besar (big data). Model dan rekomendasi strategi dari penelitian ini diharapkan memberikan kontribusi bagi perbankan Syariah di Indonesia untuk tumbuh lebih baik secara berkelanjutan, sehingga memberikan kontribusi yang lebih besar bagi perekonomian Indonesia.
================================================================================================================================
Compared to the total banking industry, Islamic banking in Indonesia has smaller assets, higher non-performing financing (NPF), a lower return on assets (ROA), and a higher dependence on financing. Theoretically, Islamic banking is directly related to sustainable finance, as it utilizes maqasid sharia as a guiding principle, highlighting the importance of this research in addressing global issues. Although predictive and prescriptive analytics have been used in various fields, no references have been found in the formulation of banking financing strategies. Using monthly banking data from the Indonesian Financial Services Authority and financial data from the Indonesian Central Bank from 2015 to 2023, this study proposes a new combination of linear programming (LP) model for revenue optimization, predictive analytics using the Markov switching vector autoregressive model (MS-VAR) method accompanied by a comparison with vector autoregressive (VAR) and machine learning (ML), and financial analysis for revenue and profitability simulation. The study also conducts an empirical analysis of the resilience of Islamic banking to crises and support for sustainable finance, which is a global initiative through the Sustainable Development Goals (SDGs). Optimization results show Islamic banking supports sustainable finance. If implemented, it will result in a 32% ROA increase in 2024. Islamic banking’s profitability aligns with the sustainability agenda, demonstrating greater resilience but less anticipatory of crises compared to the total banking industry. The LP model is the basis for optimizing forecasting for 2024-2026 with MS-VAR analysis combined with VAR or ML analysis. This strategy is projected to result in an average annual increase in ROA of 25% during 2024-2026, which exceeds the average ROA growth of the total banking industry. This study also recommends that Islamic banking focus on micro, small, and medium enterprises (MSMEs) and consumer financing, supported by focusing on several economic sectors, as they will give significant contributions to the profitability increase. This study shows that the recommended combined model is very suitable for implementation by Islamic banking. Through a feedback mechanism with financial analysis to form a prescriptive analytics model, this model produces a reliable Islamic financing strategy that can accommodate changes in economic conditions. Further research and implementation of the results of this study can be used with data for specific banks, which are usually in the form of daily data and big data. The model and strategic recommendations from this research are expected to contribute to Islamic banking in Indonesia to grow better and support sustainability practices, thus providing a greater contribution to the Indonesian economy.
| Item Type: | Thesis (Doctoral) |
|---|---|
| Uncontrolled Keywords: | Markov switching model, predictive analytics, perbankan Syariah, strategi perbankan, sustainable finance, banking strategy, Islamic banking, Markov switching model, predictive analytics, sustainable finance |
| Subjects: | H Social Sciences > H Social Sciences (General) > H61.4 Forecasting in the social sciences H Social Sciences > HA Statistics > HA30.3 Time-series analysis H Social Sciences > HG Finance |
| Divisions: | Interdisciplinary School of Management and Technology (SIMT) > 66105-Doctor of Technology Management (DMT) |
| Depositing User: | Aznovri Kurniawan |
| Date Deposited: | 23 Jul 2025 03:55 |
| Last Modified: | 23 Jul 2025 03:55 |
| URI: | http://repository.its.ac.id/id/eprint/120738 |
Actions (login required)
 |
View Item |