Working Paper: CEPR ID: DP10970
Authors: Sokbae Lee; Bernard Salani
Abstract: Multivalued treatment models have only been studied so far under restrictive assumptions: ordered choice, or more recently unordered monotonicity. We show how marginal treatment effects can be identified in a more general class of models. Our results rely on two main assumptions: treatment assignment must be a measurable function of threshold-crossing rules; and enough continuous instruments must be available. On the other hand, we do not require any kind of monotonicity condition. We illustrate our approach on several commonly used models; and we also discuss the identification power of discrete instruments.
Keywords: Discrete Choice; Identification; Monotonicity; Treatment Evaluation
JEL Codes: C14; C21
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| treatment assignment (C90) | marginal treatment effects (C32) |
| threshold-crossing rules (C24) | treatment assignment (C90) |
| continuous instruments (C36) | probability distribution of unobservables (C46) |
| treatment assignment (C90) | observed outcomes (C90) |
| unobserved variables influencing treatment (C32) | treatment assignment (C90) |
| treatment assignment (C90) | unobserved variables influencing treatment (C32) |
| marginal treatment effects (C32) | observed outcomes (C90) |