Big Data and Textual Analysis in Economics, Finance, and Politics Workshop

Key Details

Date & Time – Thursday 1 June, 10.30am – 4.00pm

Location – Room 210, Social Sciences Building A02, The University of Sydney, Camperdown NSW 2006

Big Data and Textual Analysis in Economics, Finance, and Politics Programme

Register Here (Closing date – Friday, 26 May)

Background

There has been a rapid growth of interest in big data, machine learning and textual analysis in economics, finance, and politics. This workshop features four cutting-edge studies, focusing on applications related to government behavior, quantifying the impacts of return migration, and the role of ideology in the collective construction of knowledge on the internet. It also features a keynote lecture on “Alternative Data in Economics.”

Presentations

Dr Ashani Amarasinghe utilizes “big data” to study public responses to resource windfalls across a number of countries. Combining high frequency event data with temporally granular global commodity price shocks, this paper examines how resource windfalls drive national and international political relationships.

Dr Sibo Liu applies textual analysis methods to analyse the content of local government financial disclosures. He explores how government financial disclosures differ from corporate disclosures, tests the predictive power of the cost of borrowing, economic growth, and financing and investment behaviors of local governments, and finally, the speed with which financial markets react (or fail to react) to the information in these reports. He concludes with recommendations for government financial reporting.

In our keynote lecture, Professor Paul Raschky of Monash University, a global-leading expert on big data analysis in economics, presents on “Alternative Data in Economics.” He will discuss examples from his extensive portfolio of research in the area, such as leveraging 1 trillion observations of end-user internet connections to study global internet behavior, studying the effects of banning ChatGPT on programmers’ productivity in Italy, and documenting the effects of internet disruptions from Myanmar to Ukraine. Professor Raschky will also discuss leveraging big data to inform and influence policymaking, and commercialising such analysis.

For more information on this presentation, please click here.

Professor Yifan Zhang leverages a dataset based on work histories from hundreds of millions of LinkedIn profiles, to study the impacts of high-skill return migration on firms in China. This study addresses fundamental questions about differences in economic development between countries and the role of human capital in economic development. It focuses on a country that has been one of the economic success stories of recent decades, enabling extensive overseas skills acquisition by its people.

Dr Francesco Bailo studies the role of ideology in the collective construction of knowledge on the internet, applying textual analysis and machine learning to Wikipedia. The collective construction of knowledge has become an increasingly important source of knowledge creation and belief formation in society, particularly on online platforms such as Wikipedia, online discussion channels such as reddit, social media, and citizen journalism. However, the ideological position of contributors is often unknown. This study predicts the ideology of Wikipedia editors, and then explores questions about how knowledge is created on Wikipedia.

For any enquiries or further information please contact the workshop organiser, Dr Russel Toth