| dc.contributor.advisor | Budiarti, Retno | |
| dc.contributor.advisor | Agustiani, Nur | |
| dc.contributor.author | Shalihah, Stefanny | |
| dc.date.accessioned | 2026-06-19T00:42:42Z | |
| dc.date.available | 2026-06-19T00:42:42Z | |
| dc.date.issued | 2026 | |
| dc.identifier.uri | http://repository.ipb.ac.id/handle/123456789/173508 | |
| dc.description.abstract | Ketidakpastian ekonomi global dan volatilitas pasar keuangan menyebabkan pergerakan return harga saham Crude Palm Oil (CPO) Indonesia menjadi kompleks dan sulit diprediksi dengan model linier sederhana. Penelitian ini bertujuan menganalisis pengaruh harga minyak dunia dan nilai tukar Rupiah terhadap return saham CPO serta membangun model peramalan menggunakan pendekatan Hybrid Regression–Nonlinear Autoregressive Neural Network (NARN). Data yang digunakan berupa deret waktu harian periode 4 Januari 2022 hingga 30 Maret 2026 dengan saham PT Astra Agro Lestari Tbk sebagai proksi. Analisis diawali dengan regresi linier untuk menangkap hubungan linier, kemudian residual yang masih mengandung pola dimodelkan menggunakan NARN untuk menangkap komponen nonlinier. Hasil menunjukkan bahwa return harga minyak dunia berpengaruh positif signifikan, sedangkan return nilai tukar Rupiah berpengaruh negatif signifikan terhadap return harga saham CPO. Model Hybrid Regresi–NARN mampu meningkatkan akurasi peramalan dengan nilai Root Mean Square Error (RMSE) dan Mean Absolute Error (MAE) yang rendah serta kinerja yang stabil tanpa indikasi overfitting.Temuan ini menunjukkan bahwa pendekatan hybrid efektif dalam memodelkan data keuangan yang kompleks dan relevan untuk analisis risiko serta pengambilan keputusan investasi. | |
| dc.description.abstract | Global economic uncertainty and financial market volatility make the movement of Crude Palm Oil (CPO) stock returns in Indonesia increasingly complex and difficult to predict using simple linear models. This study aims to analyze the effect of world oil prices and the Rupiah exchange rate on CPO stock returns and to develop a forecasting model using a Hybrid Regression–Nonlinear Autoregressive Neural Network (NARN) approach. The data used consist of daily time series from January 4, 2022, to March 30, 2026, with PT Astra Agro Lestari Tbk serving as a proxy for Indonesian CPO stocks. The analysis begins with linear regression to capture linear relationships, followed by modeling the residuals using NARN to account for nonlinear patterns. The results show that world oil price returns have a significant positive effect, while Rupiah exchange rate returns have a significant negative effect on CPO stock returns. The hybrid regression–NARN model improves forecasting accuracy, as indicated by lower Root Mean Square Error (RMSE) and Mean Absolute Error (MAE), with stable performance and no indication of overfitting. These findings suggest that the hybrid approach is effective in modeling complex financial data and is relevant for risk analysis and investment decision-making. | |
| dc.description.sponsorship | | |
| dc.language.iso | id | |
| dc.publisher | IPB University | id |
| dc.title | Peramalan Return Harga Saham CPO Indonesia dengan Pendekatan Hybrid Regression-Nonlinear Autoregressive Neural Network | id |
| dc.title.alternative | Forecasting Indonesian Crude Palm Oil (CPO) Stock Returns Using a Hybrid Regression–Nonlinear Autoregressive Neural Network Approach | |
| dc.type | Skripsi | |
| dc.subject.keyword | hybrid regression | id |
| dc.subject.keyword | minyak sawit | id |
| dc.subject.keyword | Crude palm oil | id |
| dc.subject.keyword | Neural Network | id |
| dc.subject.keyword | peramalan | id |
| dc.subject.keyword | return saham | id |
| dc.subject.keyword | metode hybrid | id |