D
Dooruj Rambaccussing
Researcher at University of Dundee
Publications - 21
Citations - 124
Dooruj Rambaccussing is an academic researcher from University of Dundee. The author has contributed to research in topics: Present value & Dividend. The author has an hindex of 6, co-authored 21 publications receiving 101 citations. Previous affiliations of Dooruj Rambaccussing include University of Exeter.
Papers
More filters
Journal ArticleDOI
Forecasting with social media: evidence from tweets on soccer matches
TL;DR: The authors used a microblogging dictionary to analyze the content of tweets and found that the aggregate tone of Tweets contains significant information not in betting prices, particularly in the immediate aftermath of goals and red cards.
Journal ArticleDOI
Forecasting with news sentiment: Evidence with UK newspapers
TL;DR: This article investigated the performance of newspapers for forecasting inflation, output and unemployment in the United Kingdom and found that the economic policy content reported in popular printed media can improve on existing point forecasts.
Journal ArticleDOI
A test of the long memory hypothesis based on self-similarity
TL;DR: In this paper, a new test of true versus spurious long memory, based on logperiodogram estimation of the long memory parameter using skip-sampled data, was developed, and implemented using the bootstrap, with the distribution under the null hypothesis approximated using the sieve-autoregression to approximate short-run dependence following fractional differencing.
Posted Content
Using Social Media to Identify Market Inefficiencies: Evidence from Twitter and Betfair
TL;DR: In this article, the authors analyse 13.8m posts on Twitter and high-frequency betting data from Betfair concerning English Premier League soccer matches in 2013/14 and find that the Tweets of certain journalists, and the tone of all Tweets, contain fundamental information not revealed in betting prices.
Journal ArticleDOI
True versus Spurious Long Memory in Cryptocurrencies
TL;DR: This paper finds that long memory is mostly rejected in returns, and the estimated memory parameters show that volatility is persistent, and when volatility is measured by log range, it is borderline nonstationary.