PhD Proposal -- Xinyi Zhang

Tuesday, September 25, 2018 - 9:00am
HFH 1132
Modeling Financial Behavior Online
Ben Zhao (Co-Chair), Heather Zheng (Co-Chair), William Wang

Financial behavior has long been a topic of interest in the fields of economics, psychology, and neuroscience. However, most studies on financial behavior are either studying macro trends or are in the forms of small-scale studies. This is due to the sensitive nature of financial records, which leads to a lack of large-scale dataset on financial activities. The landscape has been slowly changing. Since the emergence of e-commerce websites and the fast adoption of mobile wallets, more and more payments are conducted online, and even shared publicly.

These advancements provide rich opportunities for data-driven researches into financial behaviors. My research focuses on using data analysis and modeling techniques to extract diverse patterns in online financial behavior and accurately predict future behavior. I collect financial behavior traces from large online platforms like Venmo and DealDash. First, I will talk about how financial activities are different from traditional social activities by looking at the P2P payment platform Venmo, showing how financial activities of vastly distinct nature coexist in the platform. And then I will show how we are able to identify different behavior clusters and user communities in Venmo using clustering and community detection techniques. Second, I will discuss how we are able to predict users' future bids on the penny auction site DealDash by modeling bidding sequences.

Everyone welcome!