We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
Online ordering is currently unavailable due to technical issues. We apologise for any delays responding to customers while we resolve this. For further updates please visit our website: https://www.cambridge.org/news-and-insights/technical-incident
Drew Dimmery (New York University)
Essays on Machine Learning and Causal Inference with Application to Nonprofits
Citation:
Dimmery addresses an interesting and important substantive issue: nonprofit organizations inject a huge amount of funding into the political process through issue advocacy, yet we know relatively little about how they do so or what motivates individuals to participate on their behalf e.g., by donating money or forwarding a tweet. Dimmerys dissertation research involves a number of projects related to this issue. However, the main methodological innovation is in applying a multi-armed bandit approach to experimental design in order to allow for adaptive treatment regimes. This is particularly useful in cases like Dimmerys, where there are many possible treatments (e.g., types of appeals) under investigation. Dimmerys method allows for the experiment to be altered in real time, with no human intervention, based on observed responses. This has the potential to greatly reduce sample sizes needed to determine which of many possible treatments have a statistically significant effect.