We're excited to share a critical launch for our US microsimulation: the Enhanced Current Population Survey. This microdata integrates information from the IRS Public Use File and calibrates the results against published administrative aggregates. Our novel use of machine learning--quantile regression forests for integration and gradient descent for reweighting--ensures the highest accuracy in our nationwide predictions.

We've used this new data to analyze the impact of the Child Tax Credit and Earned Income Tax Credit provisions in the Tax Relief for American Families and Workers Act (Wyden-Smith) and President Biden's 2025 budget, and the Niskanen Center also used our platform in their new report on options for CTC reform.

This beta-launch represents the beginning of a continuous improvement cycle. We'll continue to bolster the data with additional targets, program rules, and validation, including our calibration dashboard. We also plan to extend it to more years and locales. Feel free to comment here or email us with any feedback.

Thanks,
Max Ghenis, CEO