PyData Amsterdam 2024

Raphael Tamaki

I'm a lead data scientist in the Marketing Science in Meta (Facebook). As a lead data scientist, I lead data science projects by ensuring scalability and implementation of best practices. I have experience in data pipelines and warehousing, machine-learning visibility for MLOps, stakeholder management, mentoring, and causal inference.
Among other tools and libraries, I have expertise in Airflow, XGBoost, CausalPy, PyMC, STAN, Looker, Deep Learning, Synthetic Control (for pesudo-experiments).

The speaker's profile picture

Sessions

09-19
14:40
35min
Almost Perfect: A Benchmark on Algorithms for Quasi-Experiments
Raphael Tamaki

In this presentation, we will compare four algorithms that can be used for quasi-experiments in terms of the bias and variance between predicted and actual treatment effects and the confidence/credible intervals associated with the predictions:
- Difference-in-Differences,
- Synthetic Control,
- Meta-Learners,
- Graphical Causal Models,

By the end of this lesson, attendees will understand the shortcomings and benefits of the different algorithms and be better informed about which one best suits their needs.

Rembrandt