Identification in Multiple Treatment Models under Discrete Variation (2307.06174v1)

Abstract: We develop a method to learn about treatment effects in multiple treatment models with discrete-valued instruments. We allow selection into treatment to be governed by a general class of threshold crossing models that permits multidimensional unobserved heterogeneity. Under a semi-parametric restriction on the distribution of unobserved heterogeneity, we show how a sequence of linear programs can be used to compute sharp bounds for a number of treatment effect parameters when the marginal treatment response functions underlying them remain nonparametric or are additionally parameterized.