Developing an Algorithmic Trading Bot
Do, Tri Nam (2021)
2021
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2021052110195
https://urn.fi/URN:NBN:fi:amk-2021052110195
Tiivistelmä
The idea of cloud computing was conceived a few decades ago but only until recently has it received considerable notice from the global developer communities, particularly when made remarkably more approachable and affordable for the SMEs and recreational developers with AWS Free Tier and Azure free accounts. However, the big picture of how all components of a cloud service provider work together is still quite challenging to comprehend. Therefore, the outcome which this thesis aims to achieve is to demonstrate how the most popular services provided by AWS can cooperate all together in an exemplary cloud application – an automated stock trading bot.
This thesis also provides a descriptive explanation on the fundamentals of stock investing and trading as well as some popular trading strategies, including ones with the aid of Machine Learning models to predict the price movement of certain stocks so that readers without a background in Investing or Data Science can follow through the reasoning of how the program was designed in a specific way. It is widely believed that the stock price is implausible to be predicted, so this project did not dive deep into optimizing the results but to present a suggestion as to how the use of Machine Learning can fit in the big picture.
Overall, the result of this project was within expectation, meaning that a functional prototype of the program was produced with the capability of further development and scaling up. This was accomplished by setting up a logical design of the program, gradually making each of them functional and connecting them together so they can function as a whole.
This thesis also provides a descriptive explanation on the fundamentals of stock investing and trading as well as some popular trading strategies, including ones with the aid of Machine Learning models to predict the price movement of certain stocks so that readers without a background in Investing or Data Science can follow through the reasoning of how the program was designed in a specific way. It is widely believed that the stock price is implausible to be predicted, so this project did not dive deep into optimizing the results but to present a suggestion as to how the use of Machine Learning can fit in the big picture.
Overall, the result of this project was within expectation, meaning that a functional prototype of the program was produced with the capability of further development and scaling up. This was accomplished by setting up a logical design of the program, gradually making each of them functional and connecting them together so they can function as a whole.