Methods of Microeconometric Evaluations

The projects aims to extend the microeconometric evaluation literature in several aspects. A particular concern is the use of nonparametric methods, such as matching, to avoid typical specification biases. For example various extensions of the matching approach have been proposed and successfully applied that allow for time varying starting dates of the individual treatments (such as training, active labour market policies, etc.). They are also combined with the idea of so called difference-in-difference estimation. Currently, we are also interested in the methodology for estimating treatment effects in cases where there are more than one treatment (such as the many different variants of active labour market policy). This project is also related to the other empirical evaluation projects of our group.