如何直觀理解統(tǒng)計控制
發(fā)布時間：2018-10-31 16:37
【摘要】：正多元回歸分析的魅力在于:通過保持其他干擾因素不變,讓社會科學(xué)的研究者在非實驗情景之下,完成了自然科學(xué)家在實驗室里才能做到的事情。——Wooldridge,2009正如Wooldridge所言,當(dāng)我們試圖利用觀察數(shù)據(jù)(observational data)進(jìn)行因果推論時,統(tǒng)計控制扮演著類似于實驗控制的角色。從本質(zhì)上來看,這兩類控制方法的核心思想與Mill(1848)提出的"求異法"(methodof difference)一致。Mill告訴我們,在
[Abstract]:The beauty of positive multivariate regression is that by keeping other distractions constant, social science researchers do what natural scientists can do in a laboratory, in a non-experimental setting.-Wooldridge,2009, as Wooldridge puts it. Statistical control plays a similar role to experimental control when we try to use observational data (observational data) for causal inference. In essence, the core ideas of these two types of control methods are consistent with the "difference seeking method" (methodof difference) proposed by Mill (1848). Mill tells us that
【作者單位】： 臺灣政治大學(xué);
【分類號】：C81
，
本文編號：2302897
[Abstract]:The beauty of positive multivariate regression is that by keeping other distractions constant, social science researchers do what natural scientists can do in a laboratory, in a non-experimental setting.-Wooldridge,2009, as Wooldridge puts it. Statistical control plays a similar role to experimental control when we try to use observational data (observational data) for causal inference. In essence, the core ideas of these two types of control methods are consistent with the "difference seeking method" (methodof difference) proposed by Mill (1848). Mill tells us that
【作者單位】： 臺灣政治大學(xué);
【分類號】：C81
，
本文編號：2302897
本文鏈接：http://sikaile.net/shekelunwen/shgj/2302897.html
最近更新
教材專著
