0
$\begingroup$

I have to answer some multiple choice questions about this paper and it is important that I get all of them right so I would greatly appreciate if someone could let me know if my answers are correct.

1) In the first stage, average protection against expropriation risk is regressed on settler mortality (and covariates). If the t-value of the coefficient of settler mortality is larger than 1.96, we conclude that the instrument is relevant.

True

2) We can test whether the instrument mortality influences GDP in no other way than through the endogenous explanatory variable, protection against expropriation risk, by running an auxiliary regression.

True

$\endgroup$
  • 1
    $\begingroup$ It might be sensible to provide a link to the paper. This? $\endgroup$ – Henry Nov 18 at 19:32
  • $\begingroup$ yes, that is the paper $\endgroup$ – Jagol95 Nov 18 at 19:33
  • 1
    $\begingroup$ Just a pedagogical point, but a high t-value does not mean the instrument is relevant - it means that you have failed to show that the instrument is not relevant. Or, in other words, you've failed to reject the null hypothesis that the instrument is relevant. $\endgroup$ – heh Nov 18 at 20:50