Prediksi Pergerakan Indeks Harga Saham Di Indonesia Dengan Google Trends
DOI:
https://doi.org/10.34818/eoe.v7i2.13132Abstract
Abstrak Indeks harga saham merupakan satuan nilai untuk mengetahui pergerakan harga suatu pasar saham atau sebagian pasar saham. Indeks harga saham berfungsi sebagai tolak ukur bagi para pemegang saham untuk mendapatkan keuntungan dalam berinvestasi. Oleh karena itu, diperlukan pemodelan prediksi indeks harga saham untuk membantu para pemegang saham dalam mengambil keputusan. Banyak faktor yang mempengaruhi naik turunnya suatu indeks harga saham. Salah satunya adalah faktor ekspektasi para penjual dan pembeli saham beserta perhatian orang pada informasi naik turunnya harga saham. Google Search dapat digunakan oleh para pemegang saham untuk mengetahui info tentang indeks harga saham. Grafik pencarian terhadap keywords indeks harga saham pada Google Search dapat diakses melalui layanan Google Trends. Data pencarian indeks harga saham di Google Trends merupakan cerminan perhatian orang terhadap perubahan pergerakan indeks harga saham. Pada tugas akhir ini, dilakukan pemodelan prediksi pergerakan indeks harga saham di Indonesia dengan melibatkan data Google Trends menggunakan metode Support Vector Regression (SVR). Data indeks harga saham yang digunakan adalah penutupan mingguan indeks IHSG dan LQ45. Pemodelan prediksi indeks IHSG dan LQ45 dengan melibatkan data Google Trends memiliki akurasi lebih baik dibandingkan pemodelan tanpa melibatkan data Google Trends. Namun nilai akurasi prediksi kedua model tidak signifikan berbeda. Kata kunci : indeks harga saham, prediksi, Google Trends, Support Vector Regression, IHSG, LQ45 Abstract Stock price index is a unit of value to determine the movement of a stock market or the most of the stock market price. Stock price index serves as a benchmark for shareholders to get profits in investing. Therefore, it is necessary to make a prediction model of stock price index to assist shareholders in making decisions. There are many factors that can affect the ups and downs of a stock price index. One of the factor is the expectation of stock sellers and buyers along with people's attention on the ups and downs of stock prices information. Google Search can be used by shareholders to find out info about the stock price index. Search charts of stock price index keywords on Google Search can be accessed on the Google Trends service. Stock price index search data on Google Trends is a reflection of people's attention to changes in stock price index movements. In this final project, modeling prediction of stock price index movements in Indonesia with involving Google Trends data is built using the Support Vector Regression (SVR) method. Stock price index data that we used is IDX and LQ45. Prediction model for IDX and LQ45 index with involving Google Trends data has better results than prediction model without involving Google Trends data. However, the predicted accuracy of the two models is not significantly different. Keywords: Stock price index, prediction, Google Trends, Support Vector Regression, IHSG, LQ45
Downloads
References
[1] Bursa Efek Indonesia. 2018. Informasi Umum Saham. [Online] Available at: https://www.idx.co.id/produk/saham/ [Accessed 28 November 2019].
[2] Hu H, Tang L, Zhang S, dan Wang H. 2018. Predicting the direction of stock markets using optimized neural networks with Google Trends. ScienceDirect. Neurocomputing. vol. 285 188-195
[3] Ariyo A A, Adewumi A O, dan Ayo C K. 2014. Stock Price Prediction Using the ARIMA Model. UKSimAMSS 16th International Conference on Computer Modelling and Simulation. IEEE. Durban.
[4] W. Huang, Y. Nakamori, dan S. -Y. Wang. 2004. Forecasting stock market movement direction with support vector machine. ScienceDirect. Computers & Operations Research. vol. 32(10) 2513-2522
[5] Y. Kara, M. A. Boyacioglu, dan . Ö. K. Baykan. 2011. Predicting direction of stock price index movement using artificial neural networks and support vector machines: The sample of the Istanbul Stock Exchange. ScienceDirect. Expert Systems with Applications. vol. 38(5) 5311-5319
[6] L. Khaidem, S. Saha, dan S. R. Dey. 2016. Predicting the direction of stock market prices using random forest. arXiv. Mathematical Finance. 1-20
[7] M. Qiu dan Y. Song. 2016. Predicting the Direction of Stock Market Index Movement Using an Optimized Artificial Neural Network Model. PLOS ONE. vol. 11(5) e0155133
[8] T. B. Trafalis dan H. Ince. 2000. Support Vector Machine for Regression and Applications to Financial Forecasting. Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks. IJCNN 2000. Neural Computing: New Challenges and Perspectives for the New Millennium.
[9] C.-Y. Yeh, C.-W. Huang, dan S.-J. Lee. 2011. A Multiple-Kernel Support Vector Regression Approach for Stock Market Price Forecasting. ScienceDirect. Expert Systems with Applications. vol. 38(3) 2177-2186
[10] X.-d. Zhang, A. Li, dan R. Pan. 2016. Stock Trend prediction Based on a New Status Box Method and AdaBoost Probabilistic Support Vector Machine. ScienceDirect. Applied Soft Computing. vol. 49 385-398
[11] L. Yu, Y. Zhao, L. Tang, dan Z. Yang. 2019. Online big data-driven oil consumption forecasting with Google trends. ScienceDirect. International Journal Forecasting. vol. 35 213-223
[12] Engle, R. F., & Granger, C. W. J. 1987. Co-Integration and Error Correction: Representation, Estimation, and Testing. Econometrica, vol. 55(2)
[13] K. S. Lai dan M. Lai. 1991. A Cointegration Test for Market Efficiency. Journal of Futures Markets, vol. 11(5) 567-575
[14] C. Hiemstra dam J. D. Jones. 1994. Testing for Linear and Nonlinear Granger Causality in the Stock PriceISSN : 2355-9365 e-Proceeding of Engineering : Vol.7, No.2 Agustus 2020 | Page 8457 Volume Relation. The Journal of Finance. vol. 49(5) 1639-1664
[15] H. Drucker, C. J. Burges, L. Kaufman, A. Smola, dan V. 1997. Support Vector Regression Machines. Advances in Neural Information Processing Systems. 155-161
[16] D. Basak, S. Pal, D. C. Patranabis. 2007. Support Vector Regression. Neural Information Processing. vol. 11(10)
[17] J. P. Vert, K. Tsuda, dan B. Schölkopf. 2004. A Primer on Kernel Methods. Computational Biology.
[18] Towards Data Science. 2019. Understanding Partial Auto-Correlation. [Online] Available at: https://towardsdatascience.com/understanding-partial-auto-correlation-fa39271146ac [Accessed 04 April 2020].



