The guidance note describes some recent developments of counterfactual impact evaluation methods. The three methodologies proposed are sequence analysis, dynamic matching and synthetic controls. Sequence analysis can be used in combination with other matching techniques, such as propensity score matching, when there is information on some sequences of an event happening before the treatment. In the case of active labour market policies, knowing the sequence of events and employment statuses that occur before the policy starts can be relevant information to be used in combination with matching on other relevant observable characteristics. Dynamic matching is also useful when evaluating active labour market policies, in particular in the case of ongoing programmes, i.e. programmes which can start at any point in time and where eventually all the individuals can be treated at one point. Synthetic control is an extension of difference-in-differences, and can be used to evaluate a policy which affects a whole single region or country, using a pool of non-treated regions or countries as a comparison group.
Advanced counterfactual evaluation methods: guidance document
Andrea Geraci;
2019-01-01
Abstract
The guidance note describes some recent developments of counterfactual impact evaluation methods. The three methodologies proposed are sequence analysis, dynamic matching and synthetic controls. Sequence analysis can be used in combination with other matching techniques, such as propensity score matching, when there is information on some sequences of an event happening before the treatment. In the case of active labour market policies, knowing the sequence of events and employment statuses that occur before the policy starts can be relevant information to be used in combination with matching on other relevant observable characteristics. Dynamic matching is also useful when evaluating active labour market policies, in particular in the case of ongoing programmes, i.e. programmes which can start at any point in time and where eventually all the individuals can be treated at one point. Synthetic control is an extension of difference-in-differences, and can be used to evaluate a policy which affects a whole single region or country, using a pool of non-treated regions or countries as a comparison group.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.