Estimation of Causal Peer Influence Effects

Panos Toulis, Edward Kao
; Proceedings of the 30th International Conference on Machine Learning, PMLR 28(3):1489-1497, 2013.

Abstract

The broad adoption of social media has generated interest in leveraging peer influence for inducing desired user behavior. Quantifying the causal effect of peer influence presents technical challenges, however, including how to deal with social interference, complex response functions and network uncertainty. In this paper, we extend potential outcomes to allow for interference, we introduce well-defined causal estimands of peer-influence, and we develop two estimation procedures: a frequentist procedure relying on a sequential randomization design that requires knowledge of the network but operates under complicated response functions, and a Bayesian procedure which accounts for network uncertainty but relies on a linear response assumption to increase estimation precision. Our results show the advantages and disadvantages of the proposed methods in a number of situations.

Cite this Paper


BibTeX
@InProceedings{pmlr-v28-toulis13, title = {Estimation of Causal Peer Influence Effects}, author = {Panos Toulis and Edward Kao}, booktitle = {Proceedings of the 30th International Conference on Machine Learning}, pages = {1489--1497}, year = {2013}, editor = {Sanjoy Dasgupta and David McAllester}, volume = {28}, number = {3}, series = {Proceedings of Machine Learning Research}, address = {Atlanta, Georgia, USA}, month = {17--19 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v28/toulis13.pdf}, url = {http://proceedings.mlr.press/v28/toulis13.html}, abstract = {The broad adoption of social media has generated interest in leveraging peer influence for inducing desired user behavior. Quantifying the causal effect of peer influence presents technical challenges, however, including how to deal with social interference, complex response functions and network uncertainty. In this paper, we extend potential outcomes to allow for interference, we introduce well-defined causal estimands of peer-influence, and we develop two estimation procedures: a frequentist procedure relying on a sequential randomization design that requires knowledge of the network but operates under complicated response functions, and a Bayesian procedure which accounts for network uncertainty but relies on a linear response assumption to increase estimation precision. Our results show the advantages and disadvantages of the proposed methods in a number of situations.} }
Endnote
%0 Conference Paper %T Estimation of Causal Peer Influence Effects %A Panos Toulis %A Edward Kao %B Proceedings of the 30th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2013 %E Sanjoy Dasgupta %E David McAllester %F pmlr-v28-toulis13 %I PMLR %J Proceedings of Machine Learning Research %P 1489--1497 %U http://proceedings.mlr.press %V 28 %N 3 %W PMLR %X The broad adoption of social media has generated interest in leveraging peer influence for inducing desired user behavior. Quantifying the causal effect of peer influence presents technical challenges, however, including how to deal with social interference, complex response functions and network uncertainty. In this paper, we extend potential outcomes to allow for interference, we introduce well-defined causal estimands of peer-influence, and we develop two estimation procedures: a frequentist procedure relying on a sequential randomization design that requires knowledge of the network but operates under complicated response functions, and a Bayesian procedure which accounts for network uncertainty but relies on a linear response assumption to increase estimation precision. Our results show the advantages and disadvantages of the proposed methods in a number of situations.
RIS
TY - CPAPER TI - Estimation of Causal Peer Influence Effects AU - Panos Toulis AU - Edward Kao BT - Proceedings of the 30th International Conference on Machine Learning PY - 2013/02/13 DA - 2013/02/13 ED - Sanjoy Dasgupta ED - David McAllester ID - pmlr-v28-toulis13 PB - PMLR SP - 1489 DP - PMLR EP - 1497 L1 - http://proceedings.mlr.press/v28/toulis13.pdf UR - http://proceedings.mlr.press/v28/toulis13.html AB - The broad adoption of social media has generated interest in leveraging peer influence for inducing desired user behavior. Quantifying the causal effect of peer influence presents technical challenges, however, including how to deal with social interference, complex response functions and network uncertainty. In this paper, we extend potential outcomes to allow for interference, we introduce well-defined causal estimands of peer-influence, and we develop two estimation procedures: a frequentist procedure relying on a sequential randomization design that requires knowledge of the network but operates under complicated response functions, and a Bayesian procedure which accounts for network uncertainty but relies on a linear response assumption to increase estimation precision. Our results show the advantages and disadvantages of the proposed methods in a number of situations. ER -
APA
Toulis, P. & Kao, E.. (2013). Estimation of Causal Peer Influence Effects. Proceedings of the 30th International Conference on Machine Learning, in PMLR 28(3):1489-1497

Related Material