Data and Code for: Valid t-ratio Inference for IV

Data and code for "Valid t-ratio Inference for IV"<br><br>Abstract<br><br>In the single-IV model, researchers commonly rely on &lt;i&gt;t&lt;/i&gt;-ratio-based inference, even though the literature has quantified its potentially severe large-sample distortions. Building on Stock and Yogo (2005), we introduce the &lt;i&gt;tF&lt;/i&gt; critical value function, leading to a standard error adjustment that is a smooth function of the first-stage &lt;i&gt;F&lt;/i&gt;-statistic. For one-quarter of specifications in 61 AER papers, corrected standard errors are at least 49 and 136 percent larger than conventional 2SLS standard errors at the 5-percent and 1-percent significance levels, respectively. &lt;i&gt;tF&lt;/i&gt; confidence intervals have shorter expected length than those of Anderson and Rubin (1949), whenever both are bounded.<br>