Penerapan metode spectral clustering untuk prediksi perkembangan harga komoditas pasar

Moreno, Calvin (2024) Penerapan metode spectral clustering untuk prediksi perkembangan harga komoditas pasar. Bachelor thesis, Universitas Pelita Harapan.

Preview	Text (Title) Title.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (40kB) | Preview
Preview	Text (Abstract) Abstract.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (475kB) | Preview
Preview	Text (ToC) ToC.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (920kB) | Preview
Preview	Text (Chapter1) Chapter1.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (568kB) | Preview
	Text (Chapter2) Chapter2.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (1MB)
	Text (Chapter3) Chapter3.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (1MB)
	Text (Chapter4) Chapter4.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (2MB)
	Text (Chapter5) Chapter5.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (477kB)
Preview	Text (Bibliography) Bibliography.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (184kB) | Preview
	Text (Appendices) Appendices.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (748kB)

Abstract

Pasar komoditas adalah bagian penting dalam perekonomian global yang mempengaruhi berbagai sektor industri. Komoditas seperti logam, energi, hasil pertanian, dan produk lainnya menjadi fokus utama bagi para pelaku pasar dan investor karena fluktuasi harga yang signifikan. Memprediksi pergerakan harga komoditas adalah tantangan yang kompleks karena faktor-faktor tersebut sangat dinamis dan saling terkait. Oleh karena itu, analisis yang cermat dan prediksi yang akurat sangat penting untuk membantu para pelaku pasar mengelola risiko serta membuat keputusan investasi yang tepat. Sehingga, pada penelitian ini akan diimplementasikan algoritma Spectral Clustering untuk clustering 6092 data yang didapatkan melalui website kaggle. Pergerakan harga komoditas diprediksi menggunakan model ARIMA dan analisis cluster dapat menentukan keputusan investasi. Hasil clustering menggunakan Spectral Clustering dengan menggunakan rata-rata nilai prediksi harga yang lebih detail, yaitu sebesar 2.56759681699944, menunjukkan bahwa hasil cluster 0 dengan rata-rata harga 2.57 dan cluster 1 dengan dengan rata-rata harga 2.56 merupakan harga “NATURAL GAS” yang rendah atau sedang sehingga disarankan untuk mempertimbangkan investasi. Dan, cluster 2 dengan rata-rata harga 2.67 menunjukkan harga “NATURAL GAS” sudah berada di atas ambang batas harga (harga tinggi) sehingga disarankan untuk menghindari investasi. / Commodity markets are an important part of the global economy affecting various industry sectors. Commodities such as metals, energy, agricultural products and other products are a key focus for market participants and investors due to significant price fluctuations. Predicting commodity price movements is a complex challenge as the factors are highly dynamic and interrelated. Therefore, careful analysis and accurate predictions are essential to help market participants manage risk and make the right investment decisions. So, in this research, the Spectral Clustering algorithm will be implemented to cluster 6092 data obtained through the kaggle website. Commodity price movements are predicted using the ARIMA model and cluster analysis used to determine investment decisions. The results of clustering using Spectral Clustering using a more detailed average price prediction value, which is 2.56759681699944, show that the results of cluster 0 with an average price of 2.57 and cluster 1 with an average price of 2.56 are low to medium "NATURAL GAS" prices so, it is advisable to consider investing. While cluster 2 with an average price of 2.67 shows that the price of "NATURAL GAS" is above the price threshold (high price) so it is recommended to avoid investing.

Actions (login required)

View Item