Difference-in-Differences time-variant control variable

I conduct a staggered difference-in-differences analysis across multiple groups i and multiple time periods t. Therefore, I fitted a fixed-effects model using the following equation:
Y_{it}=beta_{o}Treatment_{it} + alpha_{i} + delta_{t} + varepsilon_{it}

Treatment is dummy coded 1 for the first time period and all subsequent time periods after group i was treated, and coded 0 otherwise. Now I would like to include a time-variant control variable C into my model:
Y_{it}=beta_{o}Treatment_{it} + beta_{1}C_{it} + alpha_{i} + delta_{t} + varepsilon_{it}

I’m not sure whether C causes Y or vice versa. However, I have reason to believe that C causes the Treatment. Does this qualify C to be included as a control variable?

Cross Validated Asked by cian on November 21, 2021

0 Answers

Add your own answers!

Related Questions

Tensor Classification Models

1  Asked on September 3, 2020 by mamafoku


Maximum likelihood estimator for a discontinuous PDF

0  Asked on August 17, 2020 by probdiscr


ARDL and ECM lags

0  Asked on August 8, 2020 by php-useless


What statistical analysis to used for kinetic data with multiple groups?

1  Asked on August 5, 2020 by carlos-valenzuela


Random forest after cross validation

1  Asked on August 1, 2020 by steven-niggebrugge


Ask a Question

Get help from others!

© 2021 All rights reserved.