Abstract Details
Activity Number:
|
313
|
Type:
|
Invited
|
Date/Time:
|
Tuesday, August 5, 2014 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Section on Statistics in Marketing
|
Abstract #310654
|
|
Title:
|
What Did We Accomplish? Inferring the Causal Effect of a Market Intervention by Counterfactual Forecasting
|
Author(s):
|
Kay H. Brodersen*+ and Fabian Gallusser and Jim Koehler and Nicolas Remy and Steven L. Scott
|
Companies:
|
Google and Google and Google and Google and Google
|
Keywords:
|
causal inference ;
forecasting ;
Bayesian inference ;
time series ;
marketing ;
experimental
|
Abstract:
|
Inferring the causal effect of a market intervention (e.g., an advertising campaign) is difficult, especially when a randomized experiment was not possible. In this work, we introduce Bayesian structural time-series models that use synthetic controls (e.g., unaffected markets) to infer the temporal evolution of causal impact attributable to the intervention. Such models can be used, for example, to show how a marketing effect unfolded over time or assess whether it justified the underlying spending. Inference is based on an efficient MCMC implementation that enables near-interactive analyses. Our approach is being used in an increasing number of applications at Google, and we anticipate that it will prove useful in many analysis efforts elsewhere.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2014 program
|
2014 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Professional Development program, please contact the Education Department.
The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Copyright © American Statistical Association.