Online Program

Return to main conference page
Friday, May 31
Practice and Applications
Data and Society
Fri, May 31, 3:40 PM - 5:15 PM
Grand Ballroom J

Predicting the Success of an Crowdfunding Campaign: Spatial Location-Based Trajectory Modeling (306191)

Hong Huang, University of South Florida 
Clinton Woods, University of Northern Colorado 
*Han Yu, University of Northern Colorado 

Keywords: Space location-based trajectory modeling, Joint models, INLA, Online Crowdfunding, Gaussian Markov Random Fields

Crowdfunding has emerged as a major trend in financing various entrepreneurs including startup capital. The wide spread daily use of the internet in most parts of people’s lives has led to crowdfunding being a primarily online endeavor. Many online platforms offering different services are used for crowdfunding. Though there are different platforms, Kickstarter is currently the most popular and makes for a good representation of the online crowdfunding phenomenon. What makes for a truly successful drive to obtain donors is unknown to everybody who uses a crowdfunding service. For understanding the behavioral performance of an online crowdfunding campaign, a probabilistic representation of spatial location-based developmental trajectories for the backer count and the Dollar pledged was developed with the large, complex, and heterogeneous spatio-temporal crowdfunding data of over ninety-nine thousand observations to elaborate the effects of the spatio-temporal components on a crowdfunding campaign on the Kickstarter as well as the effects of key factors of interest on the prediction of successful Kickstarter campaigns using a joint modeling strategy.