Analiza najpopularnejših algoritmov za trgovanje z lastniškimi vrednostnimi papirji

CORETTI, MARTIN (author), Hovelja, Tomaž (mentor)

Epidemija covida-19 in posledična uvedba ukrepov za zajezitev širjenja koronavirusa je veliko delovnih mest čez noč ukinila, še posebej v storitveni panogi. Velik del javnosti je zato pritegnila možnost služenja preko spleta, a ne s proizvodnjo in prodajo izdelkov, temveč s kupovanjem in prodajanjem delnic ali kriptovalut. Zato sem se odločil, da bom v strokovni literaturi identificiral pet najpopularnejših tehničnih strategij in jih avtomatiziral v programskem jeziku Python. Vseh pet algoritmov bom nato testiral na dnevnih cenah delnic indeksa S&P 500 iz obdobja 2001-2021. Prišel sem do spoznanja, da so bile tri strategije v določenih pogojih uspešne, ostali dve pa sta začetni investicijski kapital na tak ali drugačen način izgubile. Od tod zaključek, da tehnične strategije niso večne, ampak delujejo zgolj v določenem časovnem obdobju, da je potrebno neprestano spremljati dogajanje na svetovnih trgih in ne prepustiti vseh odločitev računalniku, da je potrebno pripraviti in se do potankosti držati lastne strategije obvladovanja tveganj.

• Language: Slovenian
• Keywords: delnice, tržni algoritmi, donosnost
• Work type: Bachelor thesis/paper
• Typology: 2.11 - Undergraduate Thesis
• Organization: FRI - Faculty of Computer and Information Science
• Year: 2021
• PID: 20.500.12556/RUL-129979
• COBISS.SI-ID: 76891139
• Publication date in RUL: 09.09.2021

Secondary language

• Language: English
• Title: Analysis of the most popular algorithms for stock trading

Abstract

The epidemic of covida-19 and the subsequent introduction of measures to contain the spread of the coronavirus has led to many jobs being lost overnight, especially in the service sector. A large part of the public was therefore attracted by the possibility of making money online, not by producing and selling products, but by buying and selling shares or cryptocurrencies. I therefore decided to identify the five most popular technical strategies in the literature and automate them in the Python programming language. I then tested all five algorithms on the daily share prices of the S&P 500 index from 2001-2021. I found that three of the strategies were successful under certain conditions, while the other two lost their initial investment capital in one way or another. Hence the conclusion that technical strategies are not eternal, but only work over a certain period of time, that it is necessary to constantly monitor what is happening on the world markets and not leave all the decisions to the computer, and that it is necessary to prepare and follow one's own risk management strategy to the letter.

• Keywords: stocks, algorithmic trading, profitability