The paper "The Log of Gravity" by J.M.C. Santos Silva and Silvana Tenreyro addresses the issue of interpreting the parameters of log-linearized models estimated by ordinary least squares (OLS) as elasticities, which can be misleading in the presence of heteroskedasticity. The authors propose a pseudo-maximum likelihood (PML) estimator for constant-elasticity models, which is consistent and robust to heteroskedasticity. They illustrate this approach using the gravity equation for trade, a widely used model in international trade literature. The paper finds significant differences between estimates obtained with the proposed PML estimator and those obtained with traditional methods, even when controlling for fixed effects. The PML estimator provides a more accurate representation of the determinants of international trade, suggesting that smaller countries tend to be more open to trade, and that colonial ties and geographical proximity have less significant effects than commonly believed. The authors also discuss the limitations of other methods, such as non-linear least squares and Tobit estimators, and provide a Monte Carlo simulation to demonstrate the superior performance of the PML estimator under various conditions.The paper "The Log of Gravity" by J.M.C. Santos Silva and Silvana Tenreyro addresses the issue of interpreting the parameters of log-linearized models estimated by ordinary least squares (OLS) as elasticities, which can be misleading in the presence of heteroskedasticity. The authors propose a pseudo-maximum likelihood (PML) estimator for constant-elasticity models, which is consistent and robust to heteroskedasticity. They illustrate this approach using the gravity equation for trade, a widely used model in international trade literature. The paper finds significant differences between estimates obtained with the proposed PML estimator and those obtained with traditional methods, even when controlling for fixed effects. The PML estimator provides a more accurate representation of the determinants of international trade, suggesting that smaller countries tend to be more open to trade, and that colonial ties and geographical proximity have less significant effects than commonly believed. The authors also discuss the limitations of other methods, such as non-linear least squares and Tobit estimators, and provide a Monte Carlo simulation to demonstrate the superior performance of the PML estimator under various conditions.