Makine Öğrenmesi İle Borsa Analizi

Borsanın temel mantığı teknik analiz denilen matematiksel işlemlere, grafiklere ve bazı indikatörlere dayanmaktadır ve yatırımcılar işlemlerini bu grafik ve indikatörlerin ürettiği tahmin sonuçlarına göre gerçekleştirmektedirler. Bu projede makine öğrenimi ile geçmiş yıllara dair veriler kullanılarak bir sistem eğitilecek ve bu sistem gelecek günlerdeki bitcoin verilerini görsel hale getirip borsa hareketlerinin momentumuna göre kullanıcıya al ve sat sinyalleri üretecektir. Hedef olarak bugünün ve geleceğin değerli borsalarından birisi olan Bitcoin borsası ele alınacaktır. Doğrusal regresyon yöntemi ile Bitcoinin günlük grafikte en yüksek, en düşük, hacim ve arz-talep verileri üzerinden al-sat sinyalleri üretilecektir. Bu veriler Quandl veritabanı aracılığıyla Bitfinex bitcoin alım satım borsası tarafından elde edilecektir.

Anahtar Kelimeler:

Doğrusal regresyon, Bitcoin, makine öğrenimi, Quandl veritabanı, Borsa

Stock Market Analysis with Machine Learning

The basic logic of the stock market is based on mathematical operations called technical analysis, graphics and some indicators. Investors perform their transactions according to the forecast results produced by these charts and indicators. In this project, a system will be trained using machine learning and data from the past years, and this system will visualize the bitcoin data in the coming days and generate buy and sell signals for the user according to the momentum of the stock market movements. As a target, the Bitcoin stock market, which is one of the valuable stock markets of today and the future, will be discussed. With the linear regression method, buy-sell signals will be generated over the highest, lowest volume and supply-demand data on the daily chart of Bitcoin. These data will be obtained by the Bitfinex bitcoin exchange through the Quandl database.

Keywords:

Stock market, Linear regression, Bitcoin, machine learning, Quandl database,

PDF

___

Evans, C., Pappas, K., & Xhafa, F. (2013). Utilizing artificial neural networks and genetic algorithms to build an algo-trading model for intra-day foreign exchange speculation. Mathamatical and Computer Modelling(58), 1249-1266.
Deng, W., & Luo, Q. (2012). Stock Market Prediction Using Artificial Neural Networks. Advanced Engineering Forum(6-7), 1055-1060.
Kristoufek, L. (2013). Bitcoin meets google trends and Wikipedia. Scientific Reports. Volume (3), Issue 3415.
Polasik, M., & Piotrowska, A. (2015). Price fluctuations and the use of Bitcoin. İnternational Journal of Electronic Commerce, Volume (20), sayfa 9-49.
Dyhrberg, A. (2015). Bitcoin, gold and the dollar-A GARCH volatility analysis. Finance Research Letters, Volume (16), sayfa 85-92.
Chen, W., Xu, H., & Jia, L. (2021). Machine learning model for Bitcoin exchange rate prediction using economic and technology determinants. İnternational Journal of Forecasting(37), 1300-1301.
otomatik-kripto-para-alim-satim-botu-bitsgap-nedir, https://www.bitcoinhaber.net/otomatik-kripto-para-alim-satim-botu-bitsgap-nedir, son erişim 5/5/2021
trade-ideas, https://www.trade-ideas.com/ , son erişim 5/5/2021
best-ai-stock-trading-software, https://victorytale.com/best-ai-stock-trading-software/ , son erişim 15/5/2021
liberated stock trader, https://www.liberatedstocktrader.com/ai-stock-trading/ , son erişim 5/6/2021
tickeron-launches-ai-robot, https://startupsavant.com/news/tickeron-launches-ai-robot , son erişim 1/6/2021
BTCUSD-BTC-USD-Exchange-Rate, Quandl.(2021). https://www.quandl.com/data/BITFINEX/BTCUSD-BTC-USD-Exchange-Rate 17 Haziran 2021, son erişim 17/6/2021
Schmitz, J. (2020). https://towardsdatascience.com/the-beginning-of-a-deep-learning-trading-bot-part1-95-accuracy-is-not-enough-c338abc98fc2, son erişim 11/6/2021