Portfolio Allocation for Sellers in Online Advertising

TL;DR

This paper introduces a portfolio-based ad selection method that accounts for estimation uncertainty, reducing revenue variance and increasing expected revenue in online advertising markets.

Contribution

It applies portfolio optimization techniques to ad selection, producing a distribution over ads to mitigate uncertainty effects and improve revenue outcomes.

Findings

Decreases revenue variance due to estimation uncertainty.

Increases expected revenue compared to traditional methods.

Uses portfolio optimization to improve ad selection strategies.

Abstract

In markets for online advertising, some advertisers pay only when users respond to ads. So publishers estimate ad response rates and multiply by advertiser bids to estimate expected revenue for showing ads. Since these estimates may be inaccurate, the publisher risks not selecting the ad for each ad call that would maximize revenue. The variance of revenue can be decomposed into two components -- variance due to `uncertainty' because the true response rate is unknown, and variance due to `randomness' because realized response statistics fluctuate around the true response rate. Over a sequence of many ad calls, the variance due to randomness nearly vanishes due to the law of large numbers. However, the variance due to uncertainty doesn't diminish. We introduce a technique for ad selection that augments existing estimation and explore-exploit methods. The technique uses methods from portfolio optimization to produce a distribution over ads rather than selecting the single ad that maximizes estimated expected revenue. Over a sequence of similar ad calls, ads are selected according to the distribution. This approach decreases the effects of uncertainty and increases revenue.