Examining the Interaction between the Cryptocurrency Market Development and Activity on Leading Social Networks
|Název práce v češtině:
|Zkoumání Interakce mezi Vývojem Trhu s Kryptoměnami a Aktivitou na Předních Sociálních Sítích
|Název v anglickém jazyce:
|Examining the Interaction between the Cryptocurrency Market Development and Activity on Leading Social Networks
|Kryptoměny, Bitcoin, Sociální sentiment, Analýza sentimentu, Twitter, Sociální média, Kryptoměnová burza
|Klíčová slova anglicky:
|Cryptocurrencies, Bitcoin, Social sentiment, Sentiment analysis, Twitter, Social media, Cryptocurrency exchange
|Akademický rok vypsání:
|Institut ekonomických studií (23-IES)
|Vedoucí / školitel:
|Mgr. Nicolas Fanta
|skrytý - zadáno vedoucím/školitelem
|Datum a čas obhajoby:
|Místo konání obhajoby:
|Opletalova - Opletalova 26, O105, Opletalova - místn. č. 105
|Datum odevzdání elektronické podoby:
|Datum proběhlé obhajoby:
|doc. PhDr. Jozef Baruník, Ph.D.
|Seznam odborné literatury
|O Kraaijeveld, J De Smedt. 2020. “The predictive power of public Twitter sentiment for forecasting cryptocurrency prices“
Journal of International Financial Markets, Institutions and Money, volume 65
L. Ante. 2021. “How Elon Musk's Twitter Activity Moves Cryptocurrency Markets“
S. Wooley, A. Edmonds, A. Bagavathi and S. Krishnan. 2019. "Extracting Cryptocurrency Price Movements from the Reddit Network Sentiment," 18th IEEE International Conference On Machine Learning And Applications (ICMLA)
F Mai, Z Shan, Q Bai, X Wang, R H L Chiang. 2018. “How Does Social Media Impact Bitcoin Value? A Test of the Silent Majority Hypothesis“ Journal of Management Information Systems, volume 35
Abraham, Jethin; Higdon, Daniel; Nelson, John; and Ibarra, Juan. 2018. "Cryptocurrency Price Prediction Using Tweet Volumes and Sentiment Analysis," SMU Data Science Review: Vol. 1 : No. 3 , Article 1.
D Shen, A Urquhart, P Wang, 2019. “Does twitter predict Bitcoin?“ Economics Letters, volume 174
|Předběžná náplň práce v anglickém jazyce
|Research question and motivation
The main research question that I intend to study is whether it is possible to describe the development of the cryptocurrency market using data from social media and if so, then how precisely.
Reguralry, millions of people are locked to their devices constantly watching the cryptocurrency market, generally with the intensions of money earnings, while every time registering something new and unexpected. To better understand how and what causes some of the unexpected market changes, many found the reasons on social networks, where the news usually appears faster than anywhere else as well as they are free to be shared in matter of seconds.
It is argued that many people decisions around the cryptocurrencies are influenced by what they see on the internet, especially on social networks. This is not only due to the fact that social media are usually the fastest to inform about sudden information and news, but also, they have many users. It was proven before, that the volume of tweets (the messages on social media Twitter) is a predictor of price direction of the most traded cryptocurrencies (Abraham, Higdon, Nelson, Ibarra, 2018). Those social networks are then full of people who either represent market subjects or are somehow involved in the discussions about cryptocurrencies, either because they like or share some for them interesting post or they write one of their own. Some of them, typically famous and popular people, can with one post hit the market in such a fashion that the price of cryptocurrency changes its value by tens of percent (Ante, 2021). In this thesis, I intend to measure and quantify the impact of the social media activity on both previously mentioned social networks on the actual price, volume, along with other features from the online cryptocurrency exchange Binance.
Studying the cryptocurrency market is important, because, even though cryptocurrencies are still despised by many, and rightfully being named as way too risky as an investment or to be considered as a relevant currency. Yet many countries, such as El Salvador, are slowly starting to see them as an opportunity, or as a future of transactions. This growing interest encourages the need for further research of the cryptocurrency market, which more and more academics from around the world are resorting to. Among other things, it was discovered (Mai, Shan, Bai, Wang, Chiang, 2018) that social media, especially Twitter (Shen, Urquhart, Wang, 2019) have significant impact on the cryptocurrency market.
My thesis builds upon previous studies and expand them with additional factors, such as adding data from social network Reddit, proven to be considerable (Wooley, Edmonds, Bagavathi, Krishnan, 2019) and focusing on more cryptocurrencies. Since currently exists a large number of cryptocurrencies, I will only examine the influence on a selected sample of cryptocurrencies with the highest market capitalization and/or those that are expected to potentially show the most interesting and relevant results, selected by the system further described in the thesis. The cryptocurrencies on which the thesis focuses are Bitcoin, Litecoin and Ethereum.
This thesis is aiming to have two main objectives. The first is to show how social media activity affects the cryptocurrencies market as well as to quantificate to what extend it is being done. The second goal is to try to create a model that can predict the behaviour of the cryptocurrencies market based on actual data.
Existing research suggests that social media have an impact on the cryptocurrency market (Kraaijeveld, De Smedt, 2020). By testing the correlation between data from social media and data from the cryptocurrency exchange, the thesis adds evidence for the previous claim.
My research also contributes by revealing the influence that social media, and specifically the activity of their users, have on certain aspects of selected cryptocurrencies. Furthermore, the models and insights will help us understand the cryptocurrency environment. A reader of the thesis will acquire a knowledge of the possible reasons of cryptocurrency market’s behaviour.
Using scraping, I obtain information about the cryptocurrencies. For each cryptocurrency, I will obtain information on the corresponding closing prices, trading volume and number of trades of each cryptocurrency from the cryptocurrency exchange Binance. Those are the main aspects on which I am going to measure the impact of social media on.
From world’s largest social networks, I will obtain data of activity concerning the chosen cryptocurrencies.
Using Granger causality, I test that the data obtained from Reddit and Twitter are useful in forecasting the development of aspects of the cryptocurrencies. Then I use RStudio to create and optimize a linear model which would predict future their development. Based on the effectiveness of the model, I conclude whether the forecasting using this method is possible.
a. why is my topic interesting
b. overview of existing research
c. brief introduction to the subject
d. how I add to existing research
e. results and what they mean
f. how is the thesis organized
Literature review and hypotheses
a. literature concerning the subject
b. evidence of role of research subject on the cryptocurrency market
c. what hypotheses will be tested and how
d. building the model, identifying significant and insignificant variables
a. relevant description of data
b. how did I obtain the data
c. description of methods that I use and why do I use them
d. why I utilize the independent and dependent variables I utilize, how they are measured
e. how I perform tests
f. how does the model work
a. rejecting / not rejecting hypotheses
b. my interpretation of the results of my research
c. using model in practice, commenting on results
a. broader interpretation of results
b. issues of obtaining the data
c. implications for practice
d. issues for further research