Tong Geng, Fangzhou Sun, Di Wu, Wei Zhou, Harikesh Nair and Zhangang Lin (2021). 

  • Describes a framework to automate bids and budgets for digital ad-campaigns bought via Real-Time Bidding (RTB). Builds resource allocation on top of modern Multi Touch Attribution (MTA) models to transparently deliver on advertiser’s campaign goals. Reports on deployment at, and evaluation of deployed system via randomized controlled trials. 
  • Comments are welcome.

Joonhyuk Yang, Navdeep Sahni and Harikesh Nair (2021). 

  • Describes a new e-commerce search algorithm that leverages information about new products’ advertising to boost their search rankings on the platform, thereby aiding the discoverability of higher quality, new products on the platform. Deployed at A field experiment shows that the deployment benefitted both the platform and its users, and suggests that the default presumption in tech that the search and advertising markets should be separated should be re-thought and be subject to more empirical testing.
  • Comments are welcome.

Carlos Carrion, Zenan Wang, Harikesh Nair, Xianghong Luo, Yulin Lei, Xiliang Lin, Wenlong Chen, Qiyu Hu, Changping Peng, Yongjun Bao and Weipeng Yang (2021). 

  • Presents a method to blend ads and content on e-commerce platforms that respects interactions and externalities and balances multiple platform objectives. Method involves a deep-learning system for click-through prediction combined with a “virtual-bid” formulation for balancing objectives that is auto-tuned from historical data. The formulation is a building block for an auction payment system such as VCG that respects externalities. Deployed on’s mobile app.
  • Comments are welcome.

Caio Waisman, Harikesh Nair, Carlos Carrion and Nan Xu (2019). (Updated: March 2021).

  • An experimental design for causal inference of advertising effects for an advertiser/DSP buying ads via real-time bidding (RTB). Online experimentation is induced by randomizing the bids submitted into the RTB auction. Randomization is implemented adaptively, via a contextual bandit, so the optimal bidding policy is also learned alongside inference of the ad-lift, enabling inference of the effect of advertising while concurrently minimizing the costs of experimentation.
  • Comments are welcome.

Anna Tuchman, Harikesh Nair and Pedro Gardete (2018). Quantitative Marketing and Economics,16(2), pp. 111-174.

Treats advertising consumed by agents as the outcome of a deliberate choice, implying a role for the demand for ads that is co-determined with the demand for products. Assesses complementarities in such joint consumption between products and advertising using TV ad consumption and product purchase data. Develops a structural model of joint demand for products and ads, and uses it to assess counterfactual targeting under “addressable TV” scenarios.

Oystein Daljord, Sanjog Misra and Harikesh Nair (2016). Journal of Marketing Research, April, 55(2). pp. 161-182. (Sept 2015 version).

Main point: When contracts cannot be fully heterogenous, firms gain from choosing both agents and incentives jointly. Empirically, the ability to choose agents seems to mitigate significantly the loss in incentives from the restriction to uniform contracts. This may explain the continued prevalence of uniform contracts in real-world settings.

Scott K. Shriver, Harikesh Nair and Reto Hofstetter (2013). Management Science, 59(6), June, 1425-43. (June 2012 version).

  • Measures feedback effects between user generated content and social tie formation in an online social network, using a bounds-based strategy to account for endogenous network formation.
  • Technical Appendix: A survey to understand users’ motivations to post content and to form ties online

Paul Ellickson, Sanjog Misra and Harikesh Nair (2012). Journal of Marketing Research, (Lead Article), 49(6), Dec, pp. 750-772.

  • Investigates dynamic effects of Wal-Mart entry on Supermarket pricing strategies in the US in the 1990-s. Quantifies costs and benefits to Supermarket chains of repositioning their pricing into HiLo or EDLP policies.
  • Appendix: Wal-Mart, Supermarkets and Price Format Repositioning in the Tradepress

Sanjog Misra and Harikesh Nair (2011). Quantitative Marketing and Economics, 9(3), September, pp. 211-25.

Harikesh Nair (2007). Quantitative Marketing and Economics, 5(3), 239-292.

Develops a structural model to capture demand dynamics for durable technology products. Computes life-cycle pricing policies for firms when selling to forward-looking consumers using demand estimates. Applies model to the video-game industry to quantify impact of forward-looking behavior.