| |
Establishing Causal Inference for Video-on-Demand (VoD) Content Consumption in Cable TV Services | Date:
Time:
Venue:
| | June 21, 2017, Wednesday
2:00pm - 2:30pm
Seminar Room 2.4, Level 2
School of Information Systems
Singapore Management University
80 Stamford Road
Singapore 178902
We look forward to seeing you at this research seminar. ![]()
|
|
About the Talk Media and technology are delivering content that is shaping our society. In 2016, Nielson report shows that Americans devoted about 10 hours and 39 minutes a day to screen time. As consumption of digital content evolves, so must marketing strategies. To portray this changing landscape, we have been attempting to extract business insights from a large dataset on user TV viewing behavior provided by a digital entertainment firm. With data so voluminous, we face statistical and methods challenges that are analogous to finding ‘needles of insight in a haystack of digital data’. The firm offers TV series as videos-on-demand (VoD). To stimulate purchases, first episodes typically are offered as free sampling ‘purchase teasers.’ Our objectives have been to (1) establish causal links between household sampling and their VoD series purchases, and (2) examine the effectiveness of such sampling-based strategy, only retrospectively, without having had the foresight to instantiate a controlled experiment. The data that were shared represented what was managerially possible for the firm, and is neither as long a time-series, nor as rich in terms of the variables that were captured. Moreover, it has not been managerially possible for our firm to construct a large-scale experiment, with random household assignment, or any level of control over the data-generation process. Yet, these limitations should not deter researchers from inferring causal relationships and obtaining results that are generalizable and practical. This is a pre-conference talk for the 13th Symposium on Statistical Challenges in Electronic Commerce Research (SCERC 2017) in Ho Chi Minh City, Vietnam. About the Speaker Ai Phuong HOANG (Emmy) is a PhD candidate under the Interdisciplinary Doctoral Programme in Information Systems and Marketing at Singapore Management University. She works under the guidance of Professor Robert J. Kauffman. Her research interests involve data analytics for business, consumer and social insights. She looks at how IT transforms the market for digital information and entertainment goods, and how machine-based methods can be combined with explanatory empiricism to unravel new insights from the tremendous amount of consumers’ digital traces. From 2015 to 2016, she participated in a 10-month training residency at Carnegie Mellon University. Early in 2017, she was selected as a Young Scholar by the Pacific Telecommunications Council. Emmy received her Master of Applied Information Systems from SMU in 2013. |
