Seminars 

“Young Researchers Seminar Series” is a particular seminar series organized by the Florence Center for Data Science that invites young researchers and Ph.D. students to hold a lecture about their current research or academic work. It takes place several times during the year. People are invited to attend and the registration (free of charge) is available one week before the event on the website (under the Events page).

Matt DosSantos DiSorbo

22th May 2023

Matt DosSantos DiSorbo from Harvard Business School presented a seminar on

“Starting Strong: The Impact of Early Interventions on Employee Outcomes

Abstract:

In randomized experiments with insufficient covariates, confounders can bias point estimates. We introduce rank estimates in factorial designs and discuss their relative robustness. We argue that, in many applied settings, identifying the top-ranked intervention is more critical than recovering exact point estimates. Using data from an experiment conducted at a large financial firm, our method provides evidence that interventions in the first week of an internship have the largest sustained effect on intern rating. Further, the data is suggestive that these early interventions have the greatest impact on interns eventually accepting an extended offer. This principle — intervene early — has important managerial implications.

The recording of the seminar is available here.

Riccardo Michielan

28th OF February 2023, FROM 2.30-4 PM:

Riccardo Michielan from University of Twente presented a seminar on

“Is there geometry in real networks?”
Abstract: In the past decade, many geometric network models have been developed, assuming that each vertex is associated a position in some underlying topologic space. Geometric models formalize the idea that similar vertices are naturally likely to connect. Moreover, these models are able reproduce many properties which are commonly observed in real networks. On the other hand, it is not always possible to infer the presence of geometry in real networks, if the edge connections are the only observables. The aim of this talk is to formalize a simple statistic which counts weighted triangles: this statistic discounts the triangles that are almost surely not caused by geometry. Then, using weighted triangles we will be able to elaborate a robust technique to distinguish whether real networks are embedded in a geometric space or not.

Click here to access the abstracts of the seminar

Iavor Bojinov

29TH OF JUNE 2022, FROM 4-5 PM:

Iavor Bojinov – Harvard Business School presented a seminar on

Design & Analysis of Dynamic Panel Experiments

Abstract: Over the past few years, firms have begun to transition away from the static single intervention A/B testing into dynamic experiments, where customers’ treatments can change over time within the same experiment. This talk will present the design-based foundations for analyzing such dynamic (or sequential experiments), starting with the extreme case of running an experiment on a single unit—what’s known as time-series experiments. Next, motivated by my work to understand if humans or algorithms are better at executing large financial trades, I will lay out a framework for designing and analyzing switchback experiments, a special case of time-series experiments. Then, I will explain how to extend this framework to multiple units and what happens when these units are subject to population interference (the setting where one unit’s treatment can impact another’s outcomes). Finally, if time allows, I will conclude with a brief discussion of an empirical study that leveraged over 1,000 experiments conducted at LinkedIn to quantify the additional benefits of adopting dynamic experimentation.

Dante Amengual

17th of June 2022, from 2-3 pm:

Dante Amengual – CEMFI presented a seminar on

“Hypothesis tests with a repeatedly singular information matrix

Abstract: We study score-type tests in likelihood contexts in which the nullity of the information matrix under the null is larger than one, thereby generalizing earlier results in the literature. Examples include multivariate skew-normal distributions, Hermite expansions of Gaussian copulas, purely non-linear predictive regressions, multiplicative seasonal time series models, and multivariate regression models with selectivity. Our proposal, which involves higher-order derivatives, is asymptotically equivalent to the likelihood ratio but only requires estimation under the null. We conduct extensive Monte Carlo exercises that study the finite sample size and power properties of our proposal and compare it to alternative approaches.