2 papers
stat.CO2025
Spectrally Deconfounded Random Forests
Markus Ulmer, Cyrill Scheidegger, Peter Bühlmann
We introduce a modification of Random Forests to estimate functions when unobserved confounding variables are present. The technique is tailored for high-dimensional settings with…
stat.ME2024
Ancestor regression in structural vector autoregressive models
Christoph Schultheiss, Markus Ulmer, Peter Bühlmann
We present a new method for causal discovery in linear structural vector autoregressive models. We adapt an idea designed for independent observations to the case of time series wh…