ENGG 390 Reports (MEM Students)
Degree Program
M.E.M
Year of Graduation
2019
Sponsor
Wayfair, Boston, MA
Faculty Advisor
Dave Tabors
Document Type
Report
Publication Date
Fall 9-12-2018
Abstract
In fulfillment of the ENGM 390 requirement, the internship was completed at Wayfair’s Business Intelligence department. A novel approach of using time series data modelling to produce confidence bands that could differentiate between significant and non-significant volatility in different Wayfair KPIs was used. An approach was established that allowed for the selection of models that could serve as an early flagging mechanism for significant anomalous behavior in any KPI. With deep financial implications associated with improving website health and the achievement of a proof of concept for the approach, wide spread implementation is expected across the analytic teams of Wayfair.
Level of Access
Restricted: Campus/Dartmouth Community Only Access
Dartmouth Digital Commons Citation
Khan, Umair, "Anomaly Detection in KPIs" (2018). ENGG 390 Reports (MEM Students). 14.
https://digitalcommons.dartmouth.edu/engs390/14
Restricted
Available to Dartmouth community via local IP address.
