Eray Turkel

Eray Turkel

Senior Data Scientist, Google

Publications

You can also see my publications on my Google Scholar profile.

Publications

Eray Turkel, Anish Saha, Rhett C. Owen, Greg J. Martin, and Shoshana Vasserman

(Press Coverage- Non technical) (Dataset and Code)

We develop a machine learning model to measure the investigative content of news articles. Our method combines an unsupervised document influence model with supervised classification using text data. We use our method to examine over-time and cross-sectional patterns in news production by local newspapers in the US between 2010 and 2020. We find surprising stability in the quantity of predicted investigative articles produced over most of the time period examined, but a notable decline in the last two years of the decade, corresponding to a recent wave of newsroom layoffs.

Eray Turkel

We mathematically characterize optimal auction mechanisms in environments where a regulator’s objective function takes into account both the ad revenues collected by the platforms, and other societal objectives. Based on regulations governing traditional TV advertising, we focus on settings where the regulator cares about the share of all ads allocated to a certain subset of advertisers, or the prices paid by advertisers. We then build a statistical model using data from Twitter’s political advertising database to estimate bidding and valuation distributions of real advertisers and generate simulated auctions to analyze the implications of implementing the proposed regulatory interventions. Our findings suggest that achieving favorable societal outcomes at a small revenue cost is possible through easily implementable, simple policies.

Avidit Acharya, Edoardo Grillo, Takuo Sugaya, and Eray Turkel

We build a game theoretic model of electoral campaigns as dynamic contests in which two candidates allocate their advertising budgets over time to affect their relative popularity (i.e. odds of winning), which evolves over time as a mean-reverting stochastic process. We use our characterization of the equilibrium spending path to statistically estimate the over-time rate of decay in the effectiveness of advertising in actual campaigns, using a large database of TV advertising from the US between 2000-2014.

Working Papers

Itai Ater, Adi Shany, Brad Ross, Eray Turkel, Shoshana Vasserman (In preperation)

Municipalities around the world are increasingly adopting congestion pricing policies in order to curb road traffic. Despite the growing enthusiasm for congestion pricing by policymakers, there is limited rigorous evidence that such policies are effective at reducing the number of cars on the road, and constituent concerns over inadvertent distributional consequences have slowed the adoption of these policies by years. We analyze a major field experiment testing the efficacy of congestion pricing fees in Israel using a large panel dataset. Our analysis shows that individuals cut their congestion inducing driving in response to per-Km pricing. We focus on heterogeneity in effects and implications on highway traffic density.

Yunus Can Aybas and Eray Turkel (Revise & Resubmit, last updated Dec 2022)

We study games of Bayesian persuasion where communication is coarse. This model captures interactions between a sender and a receiver, where the sender is unable to fully describe the state or recommend all possible actions. The sender always weakly benefits from more signals, as it increases their ability to persuade. However, access to a larger signal space does not always lead to more information being sent, and the receiver might prefer outcomes with coarse communication. As a motivating example, we study advertising where a larger signal space corresponds to better targeting ability for the advertiser, and show that customers may prefer less targeting. In a class of games where the sender’s utility is independent from the state, we show that an additional signal is more valuable to the sender when the receiver is more difficult to persuade. More generally, we characterize optimal ways to send information using limited signals, show that the sender’s optimization problem can be solved by searching within a finite set, and prove an upper bound on the marginal value of a signal.

Posters and Presentations

Mine Su Erturk and Eray Turkel

(MIT Conference on Digital Experimentation, November 2020)

We study a setting where a decision maker is conducting experiments in a network environment. We assume the existence of multiple analysts conducting experiments on the same network, as is the case in many online platforms. An experiment creates negative externalities on other ongoing experiments by contaminating their results. We analyze an experimenter’s decision making problem in this setting, where the goal is learning an optimal treatment regime over the network while limiting the contamination on other experimenters. We provide theoretical regret bounds and study the performance of our suggested policy through simulations.