Choosing the Best Training Programme: Is there a Case for Statistical Treatment Rules?
研究如何利用统计模型将失业者分配到不同的积极劳动力市场项目中，发现引入统计处理规则可显著缩短平均失业时长。
When treatment effects of active labour market programmes (ALMPs) are heterogeneous in an observable way across the population, the allocation of the unemployed into different programmes becomes particularly important. In this article, we present a statistical model that can be used to allocate unemployed into different ALMPs. The model presented is a duration model that uses the timing-of-events framework to identify causal effects. We compare different assignment rules, and the results suggest that a significant reduction in the average duration of unemployment may result if a statistical treatment rule is introduced.