Econometrics of cyber-investment: Analysis of cryptocurrencies and equity-crowdfunding

Ma, Xuerui (2022) Econometrics of cyber-investment: Analysis of cryptocurrencies and equity-crowdfunding. Doctoral thesis, University of East Anglia.

[thumbnail of MA Xuerui Phd Thesis Final version.pdf]
Download (6MB) | Preview


This thesis comprises three empirical studies with the common theme of cyber-investment. The first study investigates how price returns of alternative assets such as cryptocurrencies are affected by investor sentiment. The particular focus is on the Bitcoin market, and the gold market is used as a comparator. Many measures of investor sentiment are obtained from a number of online sources including Google Trends. Principal Component Analysis (PCA) is used to reduce the dimension of the sentiment proxies. We find that the price returns of Bitcoin and gold are both responsive to investor sentiment, but that they respond in very different ways. Both the second and third studies set out to predict the level of success of equity crowdfunding campaigns. We have collected data from UK equity crowdfunding platforms, including a measure of the level of success of each campaign. A key feature of this data is that it is truncated. This is because data is only available on campaigns that are successful. It is well known that ordinary least squares regression leads to inconsistent estimation when data is truncated. Because of this, truncated regression models are applied with the results that the target amount has a negative effect on the level of success, and equity provided by the company has an inverted-U-shaped effect. The third study digs deeper into the data source, which is actually panel data: for each successfully funded company, data is available on a sequence of “pitches”. The key contribution of this study is the extension of the truncated regression model to the panel setting; the panel truncated regression estimator introduced in this study is new to the econometrics literature. The third study also extends the realm of the results of the second study, by investigating the effects of the number of team members, and also by including quadratic terms which allow us to consider the characteristics of a pitch that maximise (or minimise) the level of success. Moreover, the study includes a simple form of textual analysis, in which the impact of the presence of certain words in the pitch announcement is investigated. Some words are indeed seen to have an important effect.

Item Type: Thesis (Doctoral)
Faculty \ School: Faculty of Social Sciences > School of Economics
Depositing User: Nicola Veasy
Date Deposited: 02 Nov 2023 14:53
Last Modified: 07 Nov 2023 08:46

Actions (login required)

View Item View Item