ANALISIS SENTIMEN DALAM PERAMALAN HARGA SAHAM STARBUCKS DENGAN METODE VADER DAN ARIMAX

GABRIEL OLIVIA YVONNE MANURUNG, . (2025) ANALISIS SENTIMEN DALAM PERAMALAN HARGA SAHAM STARBUCKS DENGAN METODE VADER DAN ARIMAX. Sarjana thesis, UNIVERSITAS NEGERI JAKARTA.

	Text COVER.pdf Download (2MB)
	Text BAB 1.pdf Download (1MB)
	Text BAB 2.pdf Restricted to Registered users only Download (824kB) | Request a copy
	Text BAB 3.pdf Restricted to Registered users only Download (689kB) | Request a copy
	Text BAB 4.pdf Restricted to Registered users only Download (1MB) | Request a copy
	Text BAB 5.pdf Restricted to Registered users only Download (482kB) | Request a copy
	Text DAFTAR PUSTAKA.pdf Download (938kB)
	Text LAMPIRAN.pdf Restricted to Registered users only Download (657kB) | Request a copy

Abstract

Konflik Israel–Palestina yang memanas pada 7 Oktober 2023 memicu respons global yang meluas, termasuk di media sosial. Pada 9 Oktober 2023, serikat karyawan Starbucks mengunggah pernyataan dukungan terhadap Palestina melalui platform X (Twitter), yang kemudian dijadikan dasar gugatan oleh perusahaan. Meskipun gugatan tersebut berfokus pada penggunaan hak kekayaan intelektual, publik menafsirkan tindakan itu sebagai sikap yang berseberangan dengan dukungan terhadap Palestina, sehingga memicu gerakan boikot di berbagai negara. Dampaknya terlihat pada penurunan harga saham Starbucks sebesar 1,6% selama 11 hari berturut-turut pada 4 Desember 2023 yang menjadi penurunan terpanjang sejak IPO. Dalam konteks ini, sentimen publik di media sosial, khususnya di platform X (Twitter), berperan sebagai indikator penting untuk memahami persepsi masyarakat yang berpotensi memengaruhi pergerakan harga saham. Penelitian ini bertujuan untuk menganalisis sentimen global terhadap Starbucks terkait isu boikot pro-Israel, meramalkan pergerakan harga saham menggunakan model ARIMAX, dan mengevaluasi akurasi model. Sebanyak 41.211 tweet berbahasa Inggris dengan kata kunci "boycott starbucks" dikumpulkan melalui data crawling periode 1 Oktober 2023–31 Desember 2024, sementara data harga saham diperoleh melalui library Python yfinance. Tahapan penelitian meliputi text preprocessing, pelabelan sentimen menggunakan VADER, pemilihan model ARIMAX terbaik, serta evaluasi performa model menggunakan MAPE dan MSE. Hasil analisis sentimen menunjukkan dominasi sentimen negatif (54%), diikuti sentimen positif (43,3%), dan netral (2,7%), dengan aktivitas tertinggi pada Desember 2023. Model ARIMAX(3,0,3) terpilih berdasarkan nilai AIC dan BIC terendah yaitu 637,0925 dan 674,1182, serta memenuhi uji diagnostik normalitas, independensi, dan heterokedastisitas. Peramalan dilakukan untuk 10 hari bursa aktif (2–16 Januari 2025) dengan hasil peramalan berkisar 88,62–88,86, sementara data aktual berada pada kisaran 89,78–91,98. Evaluasi peramalan menghasilkan nilai MAPE sebesar 1,97% dan MSE sebesar 3,58, menunjukkan akurasi sangat tinggi, meskipun error cenderung meningkat seiring bertambahnya horizon peramalan. Penelitian ini membuktikan bahwa integrasi analisis sentimen media sosial, open price, dan volume perdagangan melalui model ARIMAX dapat memprediksi harga saham secara signifikan. ***** The Israel-Palestine conflict that intensified on October 7, 2023, triggered a widespread global response, including on social media. On October 9, 2023, the Starbucks workers' union posted a statement of support for Palestine on platform X (Twitter), which subsequently became the basis for a lawsuit by the company. Although the lawsuit focused on intellectual property rights usage, the public interpreted the action as a stance opposed to supporting Palestine, thereby triggering a boycott movement in various countries. The impact was visible in Starbucks' stock price decline of 1.6% over 11 consecutive days on December 4, 2023, marking the longest decline since its IPO. In this context, public sentiment on social media, particularly on platform X (Twitter), serves as an important indicator for understanding public perception that potentially influences stock price movements. This research aims to analyze global sentiment toward Starbucks related to the pro-Israel boycott issue, forecast stock price movements using the ARIMAX model, and evaluate model accuracy. A total of 41,211 English-language tweets with the keyword "boycott starbucks" were collected through data crawling for the period October 1, 2023–December 31, 2024, while stock price data was obtained through the Python library yfinance. The research stages included text preprocessing, sentiment labeling using VADER, selection of the best ARIMAX model, and evaluation of model performance using MAPE and MSE. Sentiment analysis results show a dominance of negative sentiment (54%), followed by positive sentiment (43.3%), and neutral sentiment (2.7%), with peak activity occurring in December 2023. The ARIMAX(3,0,3) model was selected based on the lowest AIC and BIC values of 637.0925 and 674.1182, respectively, and satisfied the diagnostic tests for normality, independence, and heteroskedasticity. Forecasting was conducted for 10 active trading days (January 2–16, 2025) with forecast values ranging from 88.62–88.86, while actual data ranged from 89.78–91.98. The forecast evaluation yielded a MAPE value of 1.97% and MSE of 3.58, indicating exceptional accuracy, although errors tend to increase with longer forecast horizons. This research demonstrates that integrating social media sentiment analysis, open price, and trading volume through ARIMAX model can significantly predict stock prices.

Item Type:	Thesis (Sarjana)
Additional Information:	1). Dra. Widyanti Rahayu, M.Si. ; 2). Dr. Eti Dwi Wiraningsih, M.Si.
Subjects:	Sains > Matematika
Divisions:	FMIPA > S1 Matematika
Depositing User:	Users 33075 not found.
Date Deposited:	10 Feb 2026 02:36
Last Modified:	10 Feb 2026 02:36
URI:	http://repository.unj.ac.id/id/eprint/65122

Actions (login required)

View Item