-
AISTATS 20242024The synthetic control method (SCM) has become a popular tool for estimating causal effects in policy evaluation, where a single treated unit is observed. However, SCM faces challenges in accurately predicting postintervention potential outcomes had, contrary to fact, the treatment been withheld, when the pre-intervention period is short or the post-intervention period is long. To address these issues, we
-
IAAE 2023, Research Methods and Applications on Macroeconomic Forecasting2024We propose a simple yet robust framework to nowcast recession risk at a monthly frequency in both the United States and the Euro Area. Our nowcast leverages both macroeconomic and financial conditions, and is available the first business day after the reference month closes. In particular, we argue that financial conditions are not only useful to predict future downturns–as emphasized by the existing literature–but
-
2023 Conference on Digital Experimentation @ MIT (CODE@MIT)2023In this study, we aim to present an effective methodology tailored for companies interested in implement-ing adaptive experimentation in scenarios characterized by potential selection biases or endogeneity. To illustrate our approach, we begin by delving into the realm of online experimentation. Online platforms routinely conduct thousands of A/B tests annually to gain insights into the impact of user-facing
-
2023 Conference on Digital Experimentation @ MIT (CODE@MIT)2023Online A/B tests have become an indispensable tool across all the technology industry: if performed correctly, “online” experiments can inform effective decision making and product development. It should therefore not be surprising that Gupta et al. [2019] estimates that online businesses alone collectively run hundreds of thousands of experiments annually. Modern online experiments are often run in marketplaces
-
2023 Conference on Digital Experimentation @ MIT (CODE@MIT)2023Randomized Control Trials (RCTs) are widely used across Amazon to causally estimate impacts of proposed feature changes, in order to make data-driven launch decisions. A key element of experimental design is the level of randomization, and the choice often relies on the cross-unit interaction structure. For instance, in the context of advertiser experiments, a treatment may affect the outcome of control
Related content
-
December 13, 2021Amazon Research Award recipient Éva Tardos studies complex theoretical questions that have far-ranging practical consequences.
-
November 17, 2021The paper by Özer and his co-authors examines the role of earmarking and transparency on donors and their donations.
-
October 13, 2021Amazon Scholar David Card wins half the award, while academic research consultant Guido Imbens shares in the other half.
-
September 01, 2021Amazon’s scientists have developed a variety of scientific models to help customers get the most out of their membership.
-
July 26, 2021In a paper published at INFORMS in 2020, the Amazon senior principal scientist and his co-authors factored in both revenue and "the expected utility to the customer from the purchase."
-
July 20, 2021The senior economist knows what it means to pursue a career path like hers, and she’s determined to help others along the way.