Predicting bitcoin price movements using sentiment analysis: a machine learning approach

Purpose Cryptocurrencies such as Bitcoin (BTC) attracted a lot of attention in recent months due to their unprecedented price fluctuations. This paper aims to propose a new method for predicting the direction of BTC price using linear discriminant analysis (LDA) together with sentiment analysis. Design/methodology/approach Concretely, the authors train an LDA-based classifier that uses the current BTC price information and BTC news announcements headlines to forecast the next-day direction of BTC prices. The authors compare the results with a Support Vector Machine (SVM) model and random guess approach. The use of BTC price information and news announcements related to crypto enables us to value the importance of these different sources and types of information. Findings Relative to the LDA results, the SVM model was more accurate in predicting BTC next day’s price movement. All models yielded better forecasts of an increase in tomorrow’s BTC price compared to forecasting a decrease in the crypto price. The inclusion of news sentiment resulted in the highest forecast accuracy of 0.585 on the test data, which is superior to a random guess. The LDA (SVM) model with asset specific (news sentiment and asset specific) input features ranked first within their respective model classifiers, suggesting both BTC news sentiment and asset specific are prized factors in predicting tomorrow’s price direction. Originality/value To the best of the authors’ knowledge, this is the first study to analyze the potential effect of crypto-related sentiment and BTC specific news on BTC’s price using LDA and sentiment analysis.

Three Essays on the U.S. Treasury Market

<p>My thesis consists of three essays on market microstructure. Focusing on the U.S. Treasury market, I investigate several interesting research questions by using twelve years of BrokerTec order books of 2-, 5-, and 10-year on-the-run U.S. Treasury notes from January 1, 2004 to December 31, 2015, and five years of BrokerTec order books of 3-, 7- and 30-year on-the-run U.S. Treasury securities from January 1, 2011 to December 31, 2015. In the U.S. Treasury market, BrokerTec is one of the two dominant electronic communication networks (ECNs). According to my calculations by using BrokerTec order books from 2011 to 2015, the average daily trading volume of BrokerTec on-the-run U.S. Treasury securities is about 134.9 billion U.S. dollars, which accounts for about 26% of that of the total U.S. Treasury primary dealer activity. To help a wider audience better understand the importance of the research questions in the following three chapters, Chapter 1 gives a brief introduction to the U.S. Treasury market. In Chapter 2, I investigate the impact of scheduled macroeconomic news announcements on the U.S. Treasury market efficiency. To control the microstructure noise, I employ a robust method to construct market inefficiency measures. I find that the U.S. Treasury market becomes less efficient starting from five minutes before news arrivals. The finding is robust for different sample periods, macroeconomic news announcements, and market inefficiency measures. Investor heterogeneity could explain the decreased market efficiency before scheduled news announcements. In Chapter 3, I investigate the impact of workup trading protocols on the U.S. Treasury market quality. Each transaction on the lit pool opens a workup window, during which the BrokerTec trading platform continues to receive order submissions and modifications, but only matches workup orders that have the same prices. Each workup transaction starts a new counting down of the workup clock. A workup window naturally closes either after the workup times out or when a limit order is submitted at a better price. I find that the workup trading activities decrease the market quality, in aspects of market efficiency and market liquidity. In Chapter 4, I empirically examine the role of heterogeneity in traders’ beliefs and public information shocks on traders’ order submission decisions around news announcements in the U.S. Treasury market. I find that during both the pre-announcement period and the post-announcement period, the traders tend to submit more market orders and aggressive limit orders when the market uncertainty is high. I also find that the belief heterogeneity influences investors’ trading behavior and order submission strategies around news announcements. The role of the belief heterogeneity on order aggressiveness depends on the type of news, and the magnitude of the information shocks. The impact of market uncertainty and belief heterogeneity influences traders’ submission of both of the market orders and aggressive limit orders. In Chapter 5, I provide a summary on the research findings in Chapter 2, Chapter 3 and Chapter 4. I also discuss the contributions of this thesis to the literature.</p>

Three Essays on the U.S. Treasury Market

<p>My thesis consists of three essays on market microstructure. Focusing on the U.S. Treasury market, I investigate several interesting research questions by using twelve years of BrokerTec order books of 2-, 5-, and 10-year on-the-run U.S. Treasury notes from January 1, 2004 to December 31, 2015, and five years of BrokerTec order books of 3-, 7- and 30-year on-the-run U.S. Treasury securities from January 1, 2011 to December 31, 2015. In the U.S. Treasury market, BrokerTec is one of the two dominant electronic communication networks (ECNs). According to my calculations by using BrokerTec order books from 2011 to 2015, the average daily trading volume of BrokerTec on-the-run U.S. Treasury securities is about 134.9 billion U.S. dollars, which accounts for about 26% of that of the total U.S. Treasury primary dealer activity. To help a wider audience better understand the importance of the research questions in the following three chapters, Chapter 1 gives a brief introduction to the U.S. Treasury market. In Chapter 2, I investigate the impact of scheduled macroeconomic news announcements on the U.S. Treasury market efficiency. To control the microstructure noise, I employ a robust method to construct market inefficiency measures. I find that the U.S. Treasury market becomes less efficient starting from five minutes before news arrivals. The finding is robust for different sample periods, macroeconomic news announcements, and market inefficiency measures. Investor heterogeneity could explain the decreased market efficiency before scheduled news announcements. In Chapter 3, I investigate the impact of workup trading protocols on the U.S. Treasury market quality. Each transaction on the lit pool opens a workup window, during which the BrokerTec trading platform continues to receive order submissions and modifications, but only matches workup orders that have the same prices. Each workup transaction starts a new counting down of the workup clock. A workup window naturally closes either after the workup times out or when a limit order is submitted at a better price. I find that the workup trading activities decrease the market quality, in aspects of market efficiency and market liquidity. In Chapter 4, I empirically examine the role of heterogeneity in traders’ beliefs and public information shocks on traders’ order submission decisions around news announcements in the U.S. Treasury market. I find that during both the pre-announcement period and the post-announcement period, the traders tend to submit more market orders and aggressive limit orders when the market uncertainty is high. I also find that the belief heterogeneity influences investors’ trading behavior and order submission strategies around news announcements. The role of the belief heterogeneity on order aggressiveness depends on the type of news, and the magnitude of the information shocks. The impact of market uncertainty and belief heterogeneity influences traders’ submission of both of the market orders and aggressive limit orders. In Chapter 5, I provide a summary on the research findings in Chapter 2, Chapter 3 and Chapter 4. I also discuss the contributions of this thesis to the literature.</p>

Harnessing Twitter data to survey public attention and attitudes towards COVID-19 vaccines in the UK

Attitudes to COVID-19 vaccination vary considerably within and between countries. Although the contribution of socio-demographic factors to these attitudes has been studied, the role of social media and how it interacts with news about vaccine development and efficacy is uncertain. We examined around 2 million tweets from 522,893 persons in the UK from November 2020 to January 2021 to evaluate links between Twitter content about vaccines and major scientific news announcements about vaccines. The proportion of tweets with negative vaccine content varied, with reductions of 20–24% on the same day as major news announcement. However, the proportion of negative tweets reverted back to an average of around 40% within a few days. Engagement rates were higher for negative tweets. Public health messaging could consider the dynamics of Twitter-related traffic and the potential contribution of more targeted social media campaigns to address vaccine hesitancy.

Macroeconomic News and Exchange Rates: Exploring the Role of Order Flow

This article studies the effects of macroeconomic news announcements and order flow on exchange rates in Pakistan by considering both direct and indirect information channels during news announcements periods. For this purpose, it employs GARCH models by using real-time data on macroeconomic news, order flow, and exchange rates. The findings reveal that macroeconomic news directly, and indirectly affect Pak Rupee exchange rates. The results also show that the order flow drives fluctuations in Pak Rupee exchange rates indicating the role of trade signals and trading strategies of currency traders in the exchange rate determination. Hence, as part of an aggregated economic component and means of public and private information, macroeconomic news and order flow impact Pak Rupee exchange rates as an integrated determinant. When macroeconomic news strikes the foreign exchange market, it affects the decisions of market makers, influencing order flow, and then exchange rates.

Specialized News Television: Structural, Substantive and Stylistic Features (Based on Cultural News)

Specialized news broadcasting is of particular interest and it needs to be studied. The subject of the article is the content-thematic and stylistic features of news and information-analytical programs on the Kultura (Сulture) TV channel. The research is based on the method of content analysis, linguistic methods of analysis. As a result, the content and thematic features of information broadcasting in the field of culture have been revealed. The topics of the TV channel programs are diverse; they include such types of art as theater, music, literature, visual arts, cinema, etc. By topic and modality, news stories are divided into news itself, news announcements, and news retrospectives. «News of Culture with Vladislav Flyarkovsky», as constants of information and analytical television, retaines emphasis on the presenter’s strong authorial position, depth of topic coverage, a desire to comment on a phenomenon or event, to assess it, and reveal it from all sides. The communicative and stylistic design of the final program is conditioned, on the one hand, by the format of news broadcasting, on the other hand, by the theme (in the speech of journalists the vocabulary of art history, culturolog- ical and philosophical discourses prevail; restraint, intimacy, emphasizing the importance of cultural facts is combined with analyticism, emotional and evaluative nature of information presentation). The revealed features make it possible to talk about a special television format such as symbiosis of cultural and educational journalism, and information journalism.