Estimation of marginal effects for models with alternative variable transformations

margins is a powerful postestimation command that allows the estimation of marginal effects for official and community-contributed commands, with well-dened predicted outcomes (see predict). While the use of factor-variable notation allows one to easily estimate marginal effects when interactions and polynomials are used, estimation of marginal effects when other types of transformations such as splines, logs, or fractional polynomials are used remains a challenge. In this article, I describe how margins's capabilities can be extended to analyze other variable transformations using the command f able.

margins是一款功能强大的事后估计（postestimation）命令，可针对官方及社区贡献的命令估算边际效应（marginal effects），并生成定义良好的预测结果（详见predict命令）。尽管借助因子变量标记法（factor-variable notation），用户可轻松估算引入交互项与多项式项时的边际效应，但当需处理样条函数（splines）、对数变换（logs）或分式多项式（fractional polynomials）等其他类型变量变换时，边际效应的估算仍是一项挑战。本文将介绍如何借助f able命令扩展margins的功能，以实现对其他变量变换场景的分析。