This article considers average marginal effects (AME) in a panel data fixed effects logit model. Relating the identified set of the AME to an extremal moment problem, we first show how to obtain sharp bounds on the AME straightforwardly, without any optimization. Then, we consider two strategies to build confidence intervals on the AME. In the first, we estimate the sharp bounds with a semiparametric two-step estimator. The second, very simple strategy estimates instead a quantity known to be at a bounded distance from the AME. It does not require any nonparametric estimation but may result in larger confidence intervals. Monte Carlo simulations suggest that both approaches work well in practice, the second being often very competitive. Finally, we show that our results also apply to average treatment effects, the average structural functions and ordered, fixed effects logit models.

翻译：本条考虑了小组数据固定效果日志模型中的平均边际效应( AME ) 。 将已确认的 AME 组合与极端瞬间问题联系起来, 我们首先显示如何直截了当地获得对 AME 的尖锐界限, 而不作任何优化。 然后, 我们考虑两种策略来建立对 AME 的信任间隔。 首先, 我们用一个半对称的两步估测仪来估计尖锐界限。 第二, 简单的战略估计, 而不是已知的距离与 AME 的距离。 它不要求任何非对称估计, 但可能导致更大的信任间隔。 Monte Carlo 模拟显示, 这两种方法在实践中都效果良好, 第二种往往是非常有竞争力的。 最后, 我们显示我们的结果也适用于平均的治疗效果、 平均的结构功能和定序的固定效果日志模型 。