“Cluster-Robust Inference for Extremely Unbalanced Panel Data”
This paper discusses the failure of the robustness of cluster standard error under the circumstances of extremely unbalanced panel data. Under different model settings, we show the asymptotic divergence of traditional cluster standard error and introduce the new estimator with higher efficiency theoretically. We also demonstrate the robustness of the new estimators with simulation results.
“Heterogeneous Causal Effect of Job Loss on Mental Health”
The causal effect of unemployment on people’s health has been widely considered. Most studies focused on the average negative effect of job loss on physical health, such as mortality or lifestyle. Few studies about the average impact onmentalhealth. Theheterogeneous effect of unemploymenthasbeenoverlooked. This paper implements the newcausal forest method to estimate heterogeneous treatment effects.