Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/77082
Title: Cryptocurrency price analysis
Authors: Png, Javier Han Tiong
Keywords: DRNTU::Engineering::Computer science and engineering
Issue Date: 2019
Abstract: With crypocurrencies' booming popularity in recent years, people from all walks of life are now more aware of them and are investing in it. It resembles the DotCom boom in the early 2000s, with its own dramatic rise and fall in December 2017. The project starts by understanding the reasons for stronger correlation between Bitcoin and the major altcoins, followed by attempting to predict the price of Bitcoin through sentiment analysis solely on news articles. This is done using web crawling of news articles and data stripping to allow for sentiment analysis. The web crawling process was repeated at hourly intervals for a period of eight weeks. Thereafter, a lag analysis was performed and the results showed that a two-day lag had the highest correlation to the price of BTC at 0.60991. Knowing the optimal lag to execute trade was detrimental for testing our prediction because the higher the correlation, the closer the sentiment score follows the price of Bitcoin. This will improve our prediction which will yield a higher pro t. Lastly, a three-week long simulated trade test was conducted to see how much pro t an investor would yield based on the decisions made by the prediction algorithm. This simulation was extended to two other strategies, Random Investment and CoinPredictor.io. CoinPredictor.io is a third-party online service that provides cryptocurrencies price prediction. The results showed that while our prediction did not yield the highest pro t amongst the three strategies, it is still possible to use the sentiment of news articles for price prediction only if more news sources are added and a longer data collecting period is implemented.
URI: http://hdl.handle.net/10356/77082
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:SCSE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
Final Report (amended).pdf
  Restricted Access
2.39 MBAdobe PDFView/Open

Page view(s)

232
Updated on Jun 25, 2022

Download(s) 50

33
Updated on Jun 25, 2022

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.