Henrique Laurino dos Santos
Empirical methodologist in quantitative Marketing
Embedded Files
I am an associate professor of marketing with the Saïd Business School, University of Oxford. I concluded my doctorate in Marketing at the Wharton School, University of Pennsylvania, and my dissertation work covered applications of Hawkes processes to digital marketing.
My research focuses on momentum-based marketing policies, whether that be controlling wearout in digital advertising, detecting community formation in adoption networks, or parsing the interplay between new product adoptions and user reviews. I am interested in how endogenous dynamic patterns emerge among consumers, marketers and products.
Methodologically, I work on problems around Bayesian machine learning and stochastic processes. I have a long-running love affair with Hawkes processes (to the dismay of my friends and family).
You can reach me by email at henrique.laurin[0]d[0]ssant[0]s[at]sbs.[0]x.ac.uk (change the brackets for the appropriate letters and symbols)
Published works
Published works
- Laurino dos Santos, Henrique and Jonah Berger (2022), “The Speed of Stories: Semantic Progression and Narrative Success.” Journal of Experimental Psychology: General, 151(8), 1833–1842. https://doi.org/10.1037/xge0001171
In this article we use sentence-level embeddings to plot "narrative curves" for a large number of movie scripts. We then relate those to audience popularity and find common shapes of storytelling that are correlated with high scores.
Working papers (available by request)
Working papers (available by request)
- Laurino dos Santos, Henrique. “Analyzing Privacy-Restricted Clickstream Data Using Hawkes Processes.” (available at https://ssrn.com/abstract=5135798 )
My job market paper. I propose a methodology for modelling clickstream data with Hawkes processes. Although I think (and you should too!) that Hawkes processes are a great model architecture for continuous-time, repeated events data, using them in practice poses many challenges. In this article I tackle issues of model identification, interpretation, fit diagnostics, and -- most importantly -- privacy restrictions. The toughest, and most consequential, bit of methodology here is a Heckman-style correction for when cookie consent leads to intermittent timeseries truncation.
- Laurino dos Santos, Henrique and Noah Gans. “Reducing Ad Waste: Optimal Ad Scheduling with Point Processes.”
Starting from the methodology presented in my job market paper, we solve a variety of optimal dynamic advertising problems. We consider optimal policies for advertisers subject to different legal restrictions and with varying goals and planning horizons. We find that a variety of incentives, both structural and idiosyncratic, may lead to overadvertising practices.
- Laurino dos Santos, Henrique and Jehoshua Eliashberg. “Forecasting New Product Diffusion with Minimal Early-Release Data.”
Adding to the long tradition of forecasting how new movies will perform in theaters, we present a novel model that matches previous benchmarks, but with a fraction of the observation window.
Some other research in progress
Some other research in progress
- “Customer Lifetime Valuation in Dynamic Networks,” with Patrick Zhang
- “Understanding Customer Heterogeneity from Adoption Paths in Omniplatforms,” with Raghuram Iyengar, Ashis Mishra
Page updated
Google Sites
Report abuse