2
$\begingroup$

I have a model of the form $$y_i = ax_{1i} + bx_{1i}^2 + cx_{3i} + \varepsilon_i$$ Where $x_1$ is an endogenous variable. Would I need instruments for both $x_1$ and $x_1^2$ or do I only need to implement the first stage of 2SLS for $x_1$?

Thanks, Nitin

$\endgroup$
3
  • $\begingroup$ You should do both. Sometimes a natural extra one is the instrument squared. $\endgroup$
    – oda
    Commented Apr 19, 2022 at 18:59
  • $\begingroup$ Thanks, Oda. Suppose if we have an interaction term involving an endogenous variable with an exogenous one, would then also we would require an instrument for the interaction term? $\endgroup$ Commented Apr 20, 2022 at 22:03
  • $\begingroup$ Yes. See @MichaelGmeiner's answer. $\endgroup$
    – oda
    Commented Apr 21, 2022 at 8:58

1 Answer 1

2
$\begingroup$

If $Cov(x_1, \varepsilon_i)\ne 0$, it is probably the case that $Cov(x_1^2, \varepsilon)\ne 0$, and you would need an instrument for $x_1^2$.

If $z$ is your instrument for $x_1$, then as Oda said, it is natural for $z^2$ to be the instrument for $x_1^2$.

If you have an interaction term, $x_1\cdot w$, for which $Cov(w,\varepsilon)= 0$, then becasuse $Cov(x_1, \varepsilon)\ne 0$ it is probably the case that $Cov(x_1\cdot w,\varepsilon)\ne 0$ and you would need an instrument. The natural choice of an instrument is $z\cdot w$.

Note that $Cov(z,\varepsilon)=0$ and $Cov(w,\varepsilon)= 0$ does not imply that $Cov(z\cdot w,\varepsilon)=0$. Thus, instrument validity would require a separate assumption.

$\endgroup$
3
  • $\begingroup$ Thanks a lot, Michael and Oda. Could you also suggest a textbook/article where the non-linear and interaction-based endogenous variables are discussed. Will help in citing these in my paper. $\endgroup$ Commented Apr 22, 2022 at 16:00
  • $\begingroup$ tandfonline.com/doi/full/10.1080/07474938.2018.1427486 $\endgroup$ Commented Apr 25, 2022 at 8:43
  • $\begingroup$ Hi, Suppose instead of IV 2SLS, if we use the control function approach, would we then also need to estimate separate error terms for the endogenous variable, its quadratic term and interaction term and use these three error terms in the final model? $\endgroup$ Commented May 5, 2022 at 7:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.