KMATCH: Stata module for multivariate-distance and propensity-score matching
KMATCH: Stata module for multivariate-distance and propensity-score matching
复制标题
KMATCH：用于多元距离和倾向得分匹配的 Stata 模块
DOI：
10.7892/boris.113482
复制
发表时间：
2017
期刊：
影响因子：
--
通讯作者：
Ben Jann
中科院分区：
文献类型：
--
作者：
Ben Jann
KMatch根据协变量匹配已处理和未处理的观察结果，如果提供了结果变量，则基于匹配的观察结果估计治疗效果，可选地包括回归调整偏差校正。支持多变量(马哈拉诺比斯)距离匹配和倾向分数匹配，可以使用核匹配、脊线匹配或最近邻匹配。对于核匹配和脊线匹配，提出了交叉验证等几种数据驱动的带宽选择方法。该包还包括用于评估平衡和共同支持违规的各种命令。
kmatch matches treated and untreated observations with respect to covariates and, if outcome variables are provided, estimates treatment effects based on the matched observations, optionally including regression adjustment bias-correction. Multivariate (Mahalanobis) distance matching as well as propensity score matching is supported, either using kernel matching, ridge matching, or nearest-neighbor matching. For kernel and ridge matching, several methods for data-driven bandwidth selection such as cross-validation are offered. The package also includes various commands for evaluating balancing and common-support violations.