Sidharto, Brigitta Angeline (2026) Analisis Fundamental dan Prediksi Saham Indeks LQ45 dengan Metode Support Vector Regression dan Random Forest. Masters thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 6032221145-Master_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
6032221145-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (11MB) | Request a copy |
Abstract
Pasar modal merupakan sarana investasi yang dinamis, namun memiliki tingkat volatilitas yang tinggi sehingga menyulitkan investor dalam memprediksi pergerakan harga saham secara akurat. Kondisi tersebut meningkatkan risiko kerugian akibat kesalahan dalam pengambilan keputusan investasi. Salah satu instrumen investasi yang banyak diminati investor adalah saham yang tergabung dalam Indeks LQ45 Bursa Efek Indonesia karena memiliki tingkat likuiditas tinggi, kapitalisasi pasar yang besar, serta fundamental perusahaan yang relatif baik. Meskipun demikian, pergerakan harga saham pada indeks tersebut tetap dipengaruhi oleh berbagai faktor internal maupun eksternal sehingga diperlukan suatu pendekatan yang mampu mengombinasikan analisis kondisi fundamental perusahaan dengan kemampuan prediksi harga saham. Penelitian ini bertujuan mengintegrasikan analisis fundamental dan metode machine learning untuk menghasilkan rekomendasi investasi yang lebih objektif pada perusahaan yang terdaftar dalam Indeks LQ45. Analisis fundamental dilakukan menggunakan lima rasio keuangan, yaitu Earning Per Share (EPS), Price Earnings Ratio (PER), Debt to Equity Ratio (DER), Net Profit Margin (NPM), dan Return on Assets (ROA). Prediksi harga saham dilakukan menggunakan metode Support Vector Regression (SVR) dan Random Forest dengan data harga saham bulanan periode Januari 2024 hingga April 2026. Kinerja kedua model dievaluasi menggunakan Mean Absolute Percentage Error (MAPE) dan Root Mean Square Error (RMSE) untuk menentukan model prediksi terbaik. Hasil penelitian menunjukkan bahwa dari 29 perusahaan yang memenuhi kriteria penelitian, diperoleh 15 perusahaan dengan kondisi fundamental terbaik berdasarkan hasil penyaringan menggunakan lima rasio keuangan. Berdasarkan hasil evaluasi model, metode Random Forest menghasilkan tingkat akurasi prediksi yang lebih baik dibandingkan Support Vector Regression berdasarkan nilai MAPE dan RMSE yang lebih rendah. Integrasi hasil analisis fundamental dan prediksi harga saham menghasilkan rekomendasi investasi yang menunjukkan bahwa saham CPIN, ICBP, dan UNTR layak dipertimbangkan sebagai pilihan investasi karena memiliki fundamental yang kuat dan prospek pertumbuhan yang baik. Selain itu, saham INDF direkomendasikan sebagai peluang buy on weakness, sedangkan saham PGAS memiliki prospek investasi yang menarik karena didukung oleh kondisi fundamental yang baik serta potensi capital gain yang positif.
==================================================================================================================================
The capital market is a dynamic investment platform with a high level of volatility, making it difficult for investors to accurately predict stock price movements. This condition increases the risk of financial losses resulting from inappropriate investment decisions. One of the most attractive investment instruments is the stocks included in the LQ45 Index of the Indonesia Stock Exchange, which consists of companies with high liquidity, large market capitalization, and relatively strong fundamentals. Nevertheless, stock price movements within the index are influenced by various internal and external factors. Therefore, an approach that integrates fundamental analysis with stock price prediction is required to support more objective investment decisions. This study aims to integrate fundamental analysis and machine learning methods to generate investment recommendations for companies listed in the LQ45 Index. Fundamental analysis was conducted using five financial ratios, namely Earnings Per Share (EPS), Price Earnings Ratio (PER), Debt-to-Equity Ratio (DER), Net Profit Margin (NPM), and Return on Assets (ROA). Stock price prediction was performed using Support Vector Regression (SVR) and Random Forest based on monthly stock price data from January 2024 to April 2026. The performance of both models was evaluated using Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE) to determine the best prediction model. The results indicate that, among the 29 companies meeting the research criteria, 15 companies were identified as having the strongest fundamentals based on the five financial ratios. Model evaluation results demonstrate that the Random Forest method achieved higher prediction accuracy than Support Vector Regression, as indicated by lower MAPE and RMSE values. The integration of fundamental analysis and stock price prediction generated investment recommendations showing that CPIN, ICBP, and UNTR are considered the most attractive investment choices due to their strong fundamentals and promising growth prospects. Furthermore, INDF is recommended as a buy-on-weakness opportunity, while PGAS offers an attractive investment prospect supported by solid fundamentals and positive capital gain potential.
Actions (login required)
 |
View Item |