不存在多重共线性的处理效应?因果推断中的时间顺序与 Gram-Schmidt 过程
Treatment effects without multicollinearity? Temporal order and the Gram-Schmidt process in causal inference
AI总结:
本文利用回归变量的时间顺序和 Gram-Schmidt 正交化估计无偏、稳定且标准误更低的处理效应,并将系统性歧视分解为渠道效应。
摘要:
本文纳入关于回归变量时间顺序的信息,以估计正交且具有经济学可解释性的回归系数。我们建立了 Gram-Schmidt 正交化过程的新有限样本性质。所得系数无偏且稳定,并具有比普通最小二乘法更低的标准误。我们给出了系数表示处理组平均总处理效应的条件,并将模型扩展到有序回归变量组与同时回归变量组。最后,我们重新分析了两项研究,这些研究控制了具有时间顺序且共线的特征,包括种族、教育和收入。该新方法扩展了 Bohren 等人对系统性歧视的分解,使其成为特定渠道效应,并提高了显著性水平。
英文摘要:
This paper incorporates information about the temporal order of regressors to estimate orthogonal and economically interpretable regression coefficients. We establish new finite sample properties for the Gram-Schmidt orthogonalization process. Coefficients are unbiased and stable with lower standard errors than those from Ordinary Least Squares. We provide conditions under which coefficients represent average total treatment effects on the treated and extend the model to groups of ordered and simultaneous regressors. Finally, we reanalyze two studies that controlled for temporally ordered and collinear characteristics, including race, education, and income. The new approach expands Bohren et al.'s decomposition of systemic discrimination into channel-specific effects and improves significance levels.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。