Rickard Karlsson
@rickardkarlsson.bsky.social
PhD student working on causal inference and ML @ TU Delft
rickardkarlsson.com
rickardkarlsson.com
To implement this, we design an efficient two-stage algorithm that maintains valid Type 1 error rates. Compared to our previous method (HGIC) and another strong baseline, our new method achieves higher power—meaning it’s more effective at falsification.
February 19, 2025 at 4:17 PM
To implement this, we design an efficient two-stage algorithm that maintains valid Type 1 error rates. Compared to our previous method (HGIC) and another strong baseline, our new method achieves higher power—meaning it’s more effective at falsification.
When can we falsify the assumption of no unmeasured confounding in observational studies?
In our latest work, we develop a more efficient test for falsifying this assumption when we have data from multiple environments. A quick breakdown👇
In our latest work, we develop a more efficient test for falsifying this assumption when we have data from multiple environments. A quick breakdown👇
February 19, 2025 at 4:17 PM
When can we falsify the assumption of no unmeasured confounding in observational studies?
In our latest work, we develop a more efficient test for falsifying this assumption when we have data from multiple environments. A quick breakdown👇
In our latest work, we develop a more efficient test for falsifying this assumption when we have data from multiple environments. A quick breakdown👇