Work detail

Examining the Interaction between the Cryptocurrency Market Development and Activity on Leading Social Networks

Author: Bc. Jakub Doškář
Year: 2023 - winter
Leaders: Mgr. Nicolas Fanta
Work type: Bachelors
Language: English
Pages: 49
Awards and prizes:
Abstract: In this thesis, analyses are conducted to determine whether various measurements of social
media activity, including the sentiment value of posts, can be drivers or even predictors of a
change in a selected metrices of cryptocurrencies. The analyses are performed on data collected
in one-hour and fifteen-minute time intervals from the February of 2021 until the November
2022. The results of the analysis show that variability of the closing price of Bitcoin can be to
some extent explained by sentiment derivatives only. Furthermore, it was proven that sentiment
derived from social media is significant when used as a predictor of a direction of a price
change, under specific circumstances. These results oppose the previous studies, where
sentiment was not recognized significant. Moreover, it was determined that considering onehour intervals returns marginally better outcomes than in the case of shorter time intervals. The
thesis outlines the challenges researchers can face when using this technique in their work.


Česká Spořitelna


Patria Finance