Suppose I construct the variable $U_i = S_i + T_i $, and I want to estimate the equation

$$ Y_i = \alpha + \beta U_i + \gamma X_i + \epsilon _i $$

Suppose $X_i$ and $S_i$ are exogenous, but $T_i$ is endogenous.

A valid instrument for $X_i$ appears to be be $S_i$, so that the first stage estimation would be

$$ U_i = \alpha_U + \beta_U S_i + \gamma_U X_i + \eta_i $$

Since $S_i$ is used in the construction of $U_i$, it should be a relevant instrument, and since we assume that $S_i$ is exogenous, it should also satisfy the exclusion restriction.

Supposing that the above is correct, my question is whether the IV estimation of $\beta$ is the same as just estimating $\beta_S$ in the following equation:

$$Y_i = \alpha_S + \beta_S S_i + \gamma_S X_i + \xi_i $$

Thanks for your help!


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.