We use causal forests to evaluate the heterogeneous effects of the US-China trade war on China’s exports to the US. While the exports have decreased by 70% on average after the tariff, causal forests at the HS10 digit reveals that the effects range from -800% to +180%. The products that can be...

We introduce the package ddml for Double/Debiased Machine Learning (DDML) in Stata. Estimators of causal parameters for five different econometric models are supported, allowing for flexible estimation of causal effects of endogenous variables in settings with unknown functional forms and/or...

We present Causal ABM, a methodology to derive causal structures describing complex underlying behavioral phenomena. Agent-based models (ABMs) have powerful advantages for causal modeling that have not been explored sufficiently. Unlike traditional causal estimation approaches which often result...

In the present study, we propose a simple test approach based on the work of Breitung and Candelon (2006), which allows us to test for asymmetric predictability at a pre-specified frequency. The test approach can also be used to test for causality in cointegrated systems, as illustrated by...

While we have decades of evidence that economically free economies grow faster and are more productive than un-free ones, we have less knowledge about the effect of economic freedom on groups that have traditionally been disadvantaged. I study the causal effects of large and sustained jumps in...

This paper studies if workers infer from correlation about causal effects in the context of the part-time wage penalty. Differences in hourly pay between full-time and part-time workers are strongly driven by worker selection and systematic sorting. Ignoring these selection effects can lead to...

In this paper, I try to tame "Basu's elephants" (data with extreme selection on observables). I propose new practical large-sample and finite-sample methods for estimating and inferring heterogeneous causal effects (under unconfoundedness) in the empirically relevant context of limited overlap....

This paper analyses the effects of new bike lanes on traffic volume, congestion, and accidents. In order to obtain causal estimates, I exploit the quasi-random timing and location of newly built cycle lanes. Crucially, the new cycling infrastructure replaces existing car lanes thereby reducing...

This paper provides the first causal evidence on the impact of retirement on housing choices. Our empirical strategy exploits the discontinuity in the eligibility ages for state pension as an instrument for the endogenous retirement decision and controls for time-invariant individual...

Researchers have applied correlational studies to identify dominant drivers of household energy poverty. While these studies have suggested various determinants of energy poverty (such as socioeconomic, demographic, physical, institutional and cultural factors), none have utilized a causal...