更改

添加5,680字节 、 2021年6月1日 (二) 00:53
无编辑摘要
第9行: 第9行:  
【机器翻译】在观察数据的统计分析中,倾向性评分匹配是一种统计匹配技术,它试图通过计算预测接受治疗的协变量来估计治疗、政策或其他干预的效果。PSM 试图减少由于混杂变量造成的偏倚,这些变量可以通过简单地比较接受治疗的单位和没有接受治疗的单位之间的结果来估计治疗效果。保罗 · 罗森鲍姆和唐纳德 · 鲁宾在1983年介绍了这项技术。
 
【机器翻译】在观察数据的统计分析中,倾向性评分匹配是一种统计匹配技术,它试图通过计算预测接受治疗的协变量来估计治疗、政策或其他干预的效果。PSM 试图减少由于混杂变量造成的偏倚,这些变量可以通过简单地比较接受治疗的单位和没有接受治疗的单位之间的结果来估计治疗效果。保罗 · 罗森鲍姆和唐纳德 · 鲁宾在1983年介绍了这项技术。
   −
在观察数据的统计分析中,倾向性评分匹配Propensity Score Matching (PSM)是一种统计匹配技术,用来估计治疗、政策或其他干预的效果,方法是将协变量对样本“是否接受处理”的影响考虑在内。PSM试图减少由于混杂变量造成的偏倚。这些偏倚一般会在那些只对试验单元和对照单元的结果做简单对比的评估中出现。保罗·罗森鲍姆Paul R. Rosenbaum和唐纳德·鲁宾Donald Rubin在1983年介绍了这项技术。
+
在观察数据的统计分析中,倾向性评分匹配Propensity Score Matching (PSM)是一种统计匹配技术,用来估计治疗、政策或其他干预的效果,方法是将协变量对样本“是否接受处理”的影响考虑在内。PSM试图减少由于混杂变量造成的偏倚。这些偏倚一般会在那些只对处理单元和对照单元的结果做简单对比的评估中出现。保罗·罗森鲍姆Paul R. Rosenbaum和唐纳德·鲁宾Donald Rubin在1983年介绍了这项技术。
    
The possibility of bias arises because a difference in the treatment outcome (such as the [[average treatment effect]]) between treated and untreated groups may be caused by a factor that predicts treatment rather than the treatment itself. In [[randomized experiment]]s, the randomization enables unbiased estimation of treatment effects; for each covariate, randomization implies that treatment-groups will be balanced on average, by the [[law of large numbers]]. Unfortunately, for observational studies, the assignment of treatments to research subjects is typically not random. [[Matching (statistics)|Matching]] attempts to reduce the treatment assignment bias, and mimic randomization, by creating a sample of units that received the treatment that is comparable on all observed covariates to a sample of units that did not receive the treatment.
 
The possibility of bias arises because a difference in the treatment outcome (such as the [[average treatment effect]]) between treated and untreated groups may be caused by a factor that predicts treatment rather than the treatment itself. In [[randomized experiment]]s, the randomization enables unbiased estimation of treatment effects; for each covariate, randomization implies that treatment-groups will be balanced on average, by the [[law of large numbers]]. Unfortunately, for observational studies, the assignment of treatments to research subjects is typically not random. [[Matching (statistics)|Matching]] attempts to reduce the treatment assignment bias, and mimic randomization, by creating a sample of units that received the treatment that is comparable on all observed covariates to a sample of units that did not receive the treatment.
第17行: 第17行:  
【机器翻译】出现偏倚的可能性是因为治疗组和未治疗组之间治疗结果(如平均治疗效果)的差异可能是由预测治疗的因素而不是治疗本身造成的。在随机实验中,随机化可以对治疗效果进行无偏估计; 对于每个协变量,随机化意味着治疗组将按照大数定律在平均水平上达到平衡。不幸的是,对于观察性研究来说,对研究对象的治疗分配通常不是随机的。匹配试图减少处理分配偏差,并模拟随机化,通过创建一个样本单位接受的处理是可比的所有观察到的协变量的一个样本单位没有接受处理。
 
【机器翻译】出现偏倚的可能性是因为治疗组和未治疗组之间治疗结果(如平均治疗效果)的差异可能是由预测治疗的因素而不是治疗本身造成的。在随机实验中,随机化可以对治疗效果进行无偏估计; 对于每个协变量,随机化意味着治疗组将按照大数定律在平均水平上达到平衡。不幸的是,对于观察性研究来说,对研究对象的治疗分配通常不是随机的。匹配试图减少处理分配偏差,并模拟随机化,通过创建一个样本单位接受的处理是可比的所有观察到的协变量的一个样本单位没有接受处理。
   −
之所以有可能出现偏倚,是因为试验组和对照组处理结果(如平均处理效果)的差异,可能更多反映了决定样本是否接受处理的某个影响因素,而不是处理效果本身。在随机实验中,随机化选择样本可以做到对处理效果的无偏估计,根据大数定律,随机化意味着基于协变量的平均水平,平衡分配试验组和对照组。不幸的是,对于观察性研究来说,研究对象通常不是随机接受处理的。匹配就是要减少对象非随机接受处理产生的偏倚,并模拟随机试验,方法是分别从试验单元和对照单元中各选择一个样本,让它们在所有的协变量上都比较接近。
+
之所以可能出现偏倚,是因为处理组和对照组处理结果(如平均处理效果)的差异可能更多受到决定样本是否接受处理的某个因素的影响,而不是处理本身。在随机实验中,随机化选择样本可以做到对处理效果的无偏估计,根据大数定律,随机化意味着基于协变量的平均水平,平衡分配处理组和对照组。不幸的是,对于观察性研究来说,研究对象通常不是随机接受处理的。匹配就是要减少对象非随机接受处理产生的偏倚,并模拟随机试验,方法是分别从处理单元和对照单元中各选择一个样本,让它们在所有的协变量上都比较接近。
    
For example, one may be interested to know the [[Health_effects_of_tobacco#Early_observational_studies|consequences of smoking]]. An observational study is required since it is unethical to randomly assign people to the treatment 'smoking.' The treatment effect estimated by simply comparing those who smoked to those who did not smoke would be biased by any factors that predict smoking (e.g.: gender and age). PSM attempts to control for these biases by making the groups receiving treatment and not-treatment comparable with respect to the control variables.
 
For example, one may be interested to know the [[Health_effects_of_tobacco#Early_observational_studies|consequences of smoking]]. An observational study is required since it is unethical to randomly assign people to the treatment 'smoking.' The treatment effect estimated by simply comparing those who smoked to those who did not smoke would be biased by any factors that predict smoking (e.g.: gender and age). PSM attempts to control for these biases by making the groups receiving treatment and not-treatment comparable with respect to the control variables.
第25行: 第25行:  
【机器翻译】例如,人们可能有兴趣知道吸烟的后果。因为随机分配患者接受‘吸烟’治疗是不道德的,所以需要一个观察性研究简单地比较吸烟者和不吸烟者的治疗效果会受到任何预测吸烟的因素的影响(例如:。: 性别及年龄)。PSM 试图通过使接受治疗和不接受治疗的组与控制变量相比较来控制这些偏差。
 
【机器翻译】例如,人们可能有兴趣知道吸烟的后果。因为随机分配患者接受‘吸烟’治疗是不道德的,所以需要一个观察性研究简单地比较吸烟者和不吸烟者的治疗效果会受到任何预测吸烟的因素的影响(例如:。: 性别及年龄)。PSM 试图通过使接受治疗和不接受治疗的组与控制变量相比较来控制这些偏差。
   −
例如,人们想知道吸烟的后果。但是随机分配让患者“吸烟”是不道德的,所以需要做一个观察性研究。简单对比吸烟者和不吸烟者而评估的处理效果会产生偏差,它会受到任何能影响吸烟行为因素的影响(例如:性别及年龄)。PSM要做的是通过让试验组和对照组的控制变量尽量相似来达到控制这些偏差的目的。
+
例如,人们想知道吸烟的后果。但是随机分配让患者“吸烟”是不道德的,所以需要做一个观察性研究。简单对比吸烟者和不吸烟者而评估的处理效果会产生偏差,它会受到任何能影响吸烟行为因素的影响(例如:性别及年龄)。PSM要做的是通过让处理组和对照组的控制变量尽量相似来达到控制这些偏差的目的。
      第39行: 第39行:  
【机器翻译】PSM 适用于非实验环境中因果推断和简单选择偏差的情况,其中: (i)非处理对照组中与处理单元可比的单元很少; (ii)选择与处理单元类似的比较单元子集很困难,因为必须跨一组高维预处理特征进行比较。
 
【机器翻译】PSM 适用于非实验环境中因果推断和简单选择偏差的情况,其中: (i)非处理对照组中与处理单元可比的单元很少; (ii)选择与处理单元类似的比较单元子集很困难,因为必须跨一组高维预处理特征进行比较。
   −
PSM 适用于非实验环境中因果推断和简单选择偏差的情况,其中: (i)对照组与试验组中的类似的单元很少; (ii)选择与试验单元类似的对照单元子集很困难,因为必须跨一组高维预处理特征进行比较。
+
PSM 适用于非实验环境中因果推断和简单选择偏差的情况,其中: (i)对照组与处理组中的类似的单元很少; (ii)选择与处理单元类似的对照单元子集很困难,因为必须跨一组高维预处理特征进行比较。
      第48行: 第48行:  
【机器翻译】在正常的匹配中,区分治疗组和对照组的单一特征被匹配,试图使这些组更加相似。但是,如果这两个组没有实质性的重叠,那么可能会引入实质性的错误。例如,如果只将来自未经治疗的对照组的最差病例与来自治疗组的最好病例进行比较,结果可能是趋中回归,这可能使对照组看起来比实际情况更好或更糟。
 
【机器翻译】在正常的匹配中,区分治疗组和对照组的单一特征被匹配,试图使这些组更加相似。但是,如果这两个组没有实质性的重叠,那么可能会引入实质性的错误。例如,如果只将来自未经治疗的对照组的最差病例与来自治疗组的最好病例进行比较,结果可能是趋中回归,这可能使对照组看起来比实际情况更好或更糟。
   −
在正常的匹配中,区分试验组和对照组的单一特征被匹配,试图使这些组更加相似。但是,如果这两个组没有实质性的重叠,那么可能会引入实质性的错误。例如,如果只将来自未经治疗的对照组的最差病例与来自试验组的最好病例进行比较,结果可能是趋中回归,这可能使对照组看起来比实际情况更好或更糟。
+
在正常的匹配中,区分处理组和对照组的单一特征被匹配,试图使这些组更加相似。但是,如果这两个组没有实质性的重叠,那么可能会引入实质性的错误。例如,如果只将来自未经治疗的对照组的最差病例与来自处理组的最好病例进行比较,结果可能是趋中回归,这可能使对照组看起来比实际情况更好或更糟。
      第57行: 第57行:  
【机器翻译】PSM 使用了一种预测的群体成员概率---- 例如,治疗组与控制组---- 基于观察预测,通常从 Logit模型获得来创造一个反事实的群体。倾向得分可用于匹配或作为协变量,单独或与其他匹配变量或协变量。
 
【机器翻译】PSM 使用了一种预测的群体成员概率---- 例如,治疗组与控制组---- 基于观察预测,通常从 Logit模型获得来创造一个反事实的群体。倾向得分可用于匹配或作为协变量,单独或与其他匹配变量或协变量。
   −
PSM使用了一种基于观察变量预测样本落入不同分组(例如,试验组与控制组)的概率的方法来创造一个反事实的群体,通常用Logistic回归来预测概率。倾向性评分可用于匹配,也可作为协变量,可以单独使用,也可以与其他匹配变量或协变量一同使用。
+
PSM使用了一种基于观察变量预测样本落入不同分组(例如,处理组与控制组)的概率的方法来创造一个反事实的群体,通常用逻辑回归来预测概率。倾向性评分可用于匹配,也可作为协变量,可以单独使用,也可以与其他匹配变量或协变量一同使用。
      第72行: 第72行:  
*Dependent variable: ''Z'' = 1, if unit participated (i.e. is member of the treatment group); ''Z'' = 0, if unit did not participate (i.e. is member of the control group).
 
*Dependent variable: ''Z'' = 1, if unit participated (i.e. is member of the treatment group); ''Z'' = 0, if unit did not participate (i.e. is member of the control group).
   −
*因变量:''Z'' = 1, 如果参加试验(属于试验组);''Z'' = 0,如果未参加试验(属于对照组)。
+
*因变量:''Z'' = 1, 如果参与处理(属于处理组);''Z'' = 0,如果未参与处理(属于对照组)。
    
*Choose appropriate confounders (variables hypothesized to be associated with both treatment and outcome)
 
*Choose appropriate confounders (variables hypothesized to be associated with both treatment and outcome)
第89行: 第89行:  
2. 【机器翻译】检查协变量是平衡的治疗和比较组内的倾向分层。
 
2. 【机器翻译】检查协变量是平衡的治疗和比较组内的倾向分层。
   −
2. <font color="#32cd32">检查协变量在倾向性评分的分层中试验组和对照组是否平衡</font>
+
2. <font color="#32cd32">检查协变量在倾向性评分的分层中处理组和对照组是否平衡</font>
    
* Use standardized differences or graphs to examine distributions
 
* Use standardized differences or graphs to examine distributions
第102行: 第102行:  
3. 【机器翻译】根据倾向得分,将每个参与者与一个或多个非参与者进行匹配,使用以下方法之一:
 
3. 【机器翻译】根据倾向得分,将每个参与者与一个或多个非参与者进行匹配,使用以下方法之一:
   −
3. 根据倾向性评分,将每个试验参与者与一个或多个非试验参与者进行匹配,使用以下方法之一:
+
3. 根据倾向性评分,将每个参与者与一个或多个非参与者进行匹配,使用以下方法之一:
    
*[[Nearest neighbor search|Nearest neighbor matching]]
 
*[[Nearest neighbor search|Nearest neighbor matching]]
第110行: 第110行:  
*Caliper matching: comparison units within a certain width of the propensity score of the treated units get matched, where the width is generally a fraction of the standard deviation of the propensity score  
 
*Caliper matching: comparison units within a certain width of the propensity score of the treated units get matched, where the width is generally a fraction of the standard deviation of the propensity score  
   −
*卡钳匹配:用试验单元的一段宽度的倾向性评分范围选取对照单元,通常用倾向性评分的标准差的一部分确定这一宽度
+
*卡钳匹配:用处理单元的一段宽度的倾向性评分范围选取对照单元,通常用倾向性评分的标准差的一部分确定这一宽度
    
*[[Mahalanobis distance|Mahalanobis metric]] matching in conjunction with PSM
 
*[[Mahalanobis distance|Mahalanobis metric]] matching in conjunction with PSM
第136行: 第136行:  
4.【机器翻译】验证协变量是平衡的处理和对照组在匹配或加权样本
 
4.【机器翻译】验证协变量是平衡的处理和对照组在匹配或加权样本
   −
4. <font color="#32cd32">验证协变量在跨越试验组和对照组的匹配样本或加权样本中是否平衡</font>
+
4. <font color="#32cd32">验证协变量在跨越处理组和对照组的匹配样本或加权样本中是否平衡</font>
    
5. Multivariate analysis based on new sample
 
5. Multivariate analysis based on new sample
第148行: 第148行:  
*Use analyses appropriate for non-independent matched samples if more than one nonparticipant is matched to each participant
 
*Use analyses appropriate for non-independent matched samples if more than one nonparticipant is matched to each participant
   −
*如果每个试验参与者都匹配了多个非试验参与者,则应用适当的非独立匹配样本分析
+
*如果每个参与者都匹配了多个非参与者,则应用适当的非独立匹配样本分析
      第171行: 第171行:  
The basic case<ref name="Rosenbaum 1983 41–55"/> is of two treatments (numbered 1 and 0), with ''N'' [Independent and identically distributed random variables|i.i.d] subjects. Each subject ''i'' would respond to the treatment with <math>r_{1i}</math> and to the control with <math>r_{0i}</math>. The quantity to be estimated is the [[average treatment effect]]: <math>E[r_1]-E[r_0]</math>. The variable <math>Z_i</math> indicates if subject ''i'' got treatment (''Z''&nbsp;=&nbsp;1) or control (''Z''&nbsp;=&nbsp;0). Let <math>X_i</math> be a vector of observed pretreatment measurement (or covariate) for the ''i''th subject. The observations of <math>X_i</math> are made prior to treatment assignment, but the features in <math>X_i</math> may not include all (or any) of the ones used to decide on the treatment assignment. The numbering of the units (i.e.: ''i''&nbsp;=&nbsp;1,&nbsp;...,&nbsp;''i''&nbsp;=&nbsp;''N'') are assumed to not contain any information beyond what is contained in <math>X_i</math>. The following sections will omit the ''i'' index while still discussing about the stochastic behavior of some subject.
 
The basic case<ref name="Rosenbaum 1983 41–55"/> is of two treatments (numbered 1 and 0), with ''N'' [Independent and identically distributed random variables|i.i.d] subjects. Each subject ''i'' would respond to the treatment with <math>r_{1i}</math> and to the control with <math>r_{0i}</math>. The quantity to be estimated is the [[average treatment effect]]: <math>E[r_1]-E[r_0]</math>. The variable <math>Z_i</math> indicates if subject ''i'' got treatment (''Z''&nbsp;=&nbsp;1) or control (''Z''&nbsp;=&nbsp;0). Let <math>X_i</math> be a vector of observed pretreatment measurement (or covariate) for the ''i''th subject. The observations of <math>X_i</math> are made prior to treatment assignment, but the features in <math>X_i</math> may not include all (or any) of the ones used to decide on the treatment assignment. The numbering of the units (i.e.: ''i''&nbsp;=&nbsp;1,&nbsp;...,&nbsp;''i''&nbsp;=&nbsp;''N'') are assumed to not contain any information beyond what is contained in <math>X_i</math>. The following sections will omit the ''i'' index while still discussing about the stochastic behavior of some subject.
   −
基本场景<ref name="Rosenbaum 1983 41–55"/>是,有两种处理方式(分别记为1和0),''N''个[独立同分布随机变量|i.i.d]物体。每个物体''i''如果做了试验则响应是<math>r_{1i}</math>,否则响应是<math>r_{0i}</math>。被估计量是[[平均处理效应]]:<math>E[r_1]-E[r_0]</math>。变量<math>Z_i</math>指示物体''i''接受处理(''Z''&nbsp;=&nbsp;1)还是接受控制(''Z''&nbsp;=&nbsp;0)。让<math>X_i</math>代表第''i''个物体处理前观测值(或者协变量)的向量。对<math>X_i</math>的测量发生于处理前,但是<math>X_i</math>中可以不包括哪些决定是否接受处理的特征。假设对单元的编号(即:''i''&nbsp;=&nbsp;1,&nbsp;...,&nbsp;''i''&nbsp;=&nbsp;''N'')不包含任何<math>X_i</math>所包含信息之外的的信息。以下部分将省略索引''i'',同时仍会讨论某些物体的随机行为。
+
基本场景<ref name="Rosenbaum 1983 41–55"/>是,有两种处理方式(分别记为1和0),''N''个[独立同分布随机变量|i.i.d]物体。每个物体''i''如果接受了处理则响应为<math>r_{1i}</math>,否则响应为<math>r_{0i}</math>。被估计量是[[平均处理效应]]:<math>E[r_1]-E[r_0]</math>。变量<math>Z_i</math>指示物体''i''接受处理(''Z''&nbsp;=&nbsp;1)还是接受控制(''Z''&nbsp;=&nbsp;0)。让<math>X_i</math>代表第''i''个物体处理前观测值(或者协变量)的向量。对<math>X_i</math>的测量发生于处理前,但是<math>X_i</math>中可以不包括哪些决定是否接受处理的特征。假设对单元的编号(即:''i''&nbsp;=&nbsp;1,&nbsp;...,&nbsp;''i''&nbsp;=&nbsp;''N'')不包含任何<math>X_i</math>所包含信息之外的的信息。以下部分将省略索引''i'',同时仍会讨论某些物体的随机行为。
      第184行: 第184行:  
Let some subject have a vector of covariates ''X'' (i.e.: conditionally unconfounded), and some '''potential outcomes''' ''r''<sub>0</sub> and ''r''<sub>1</sub> under control and treatment, respectively. Treatment assignment is said to be '''strongly ignorable''' if the potential outcomes are [[statistical independence|independent]] of treatment (''Z'') conditional on background variables ''X''.  This can be written compactly as
 
Let some subject have a vector of covariates ''X'' (i.e.: conditionally unconfounded), and some '''potential outcomes''' ''r''<sub>0</sub> and ''r''<sub>1</sub> under control and treatment, respectively. Treatment assignment is said to be '''strongly ignorable''' if the potential outcomes are [[statistical independence|independent]] of treatment (''Z'') conditional on background variables ''X''.  This can be written compactly as
   −
设某个物体有协变量向量''X''(即:条件非混杂变量),以及一些'''潜在结果'''''r''<sub>0</sub>和''r''<sub>1</sub>分别对应着控制和处理两种情况。如果潜在结果在给定背景变量''X''的条件下独立于处理举动(''Z''),则可以说对样本接受处理的指配是'''强可忽略'''的。可简洁表述为
+
设某个物体有协变量向量''X''(即:条件无混杂),以及一些'''潜在结果‘’‘ ''r''<sub>0</sub>和''r''<sub>1</sub>分别对应着控制和处理两种情况。如果潜在结果在给定背景变量''X''的条件下独立于处理举动(''Z''),则可以说对样本是否接受处理的分配是'''强可忽略'''的。可简洁表述为
    
:<math> r_0, r_1 \perp Z \mid X </math>
 
:<math> r_0, r_1 \perp Z \mid X </math>
   −
这里<math>\perp</math>代表[[statistical independence|统计独立]].<ref name="Rosenbaum 1983 41–55"/>
+
这里<math>\perp</math>代表[[statistical independence|统计独立]].<ref name="Rosenbaum 1983 41–55"/>
    
===Balancing score===
 
===Balancing score===
 +
 +
===平衡得分===
 +
 
A '''balancing score''' ''b''(''X'') is a function of the observed covariates ''X'' such that the [[conditional probability|conditional distribution]] of ''X'' given ''b''(''X'') is the same for treated (''Z''&nbsp;=&nbsp;1) and control (''Z''&nbsp;=&nbsp;0) units:
 
A '''balancing score''' ''b''(''X'') is a function of the observed covariates ''X'' such that the [[conditional probability|conditional distribution]] of ''X'' given ''b''(''X'') is the same for treated (''Z''&nbsp;=&nbsp;1) and control (''Z''&nbsp;=&nbsp;0) units:
 +
 +
A balancing score b(X) is a function of the observed covariates X such that the conditional distribution of X given b(X) is the same for treated (Z = 1) and control (Z = 0) units:
 +
 +
平衡得分b(X)是观测协变量X的函数。在给定b(X)时,处理单元和控制单元的X有相同的条件分布:
    
:<math> Z \perp X \mid b(X).</math>
 
:<math> Z \perp X \mid b(X).</math>
   −
The most trivial function is <math> b(X) = X</math>.
+
最一般的平衡得分函数是<math> b(X) = X</math>.
    
===Propensity score===
 
===Propensity score===
 +
 +
===倾向性评分===
 +
 
A '''propensity score''' is the [[probability]] of a unit (e.g., person, classroom, school) being assigned to a particular treatment given a set of observed covariates.  Propensity scores are used to reduce [[selection bias]] by equating groups based on these covariates.
 
A '''propensity score''' is the [[probability]] of a unit (e.g., person, classroom, school) being assigned to a particular treatment given a set of observed covariates.  Propensity scores are used to reduce [[selection bias]] by equating groups based on these covariates.
    
Suppose that we have a binary treatment [[Indicator function|indicator]] ''Z'', a response variable ''r'', and background observed covariates ''X''.  The propensity score is defined as the [[conditional probability]] of treatment given background variables:
 
Suppose that we have a binary treatment [[Indicator function|indicator]] ''Z'', a response variable ''r'', and background observed covariates ''X''.  The propensity score is defined as the [[conditional probability]] of treatment given background variables:
 +
 +
A propensity score is the probability of a unit (e.g., person, classroom, school) being assigned to a particular treatment given a set of observed covariates. Propensity scores are used to reduce selection bias by equating groups based on these covariates.
 +
 +
Suppose that we have a binary treatment indicator Z, a response variable r, and background observed covariates X. The propensity score is defined as the conditional probability of treatment given background variables:
 +
 +
倾向性评分是根据协变量观测值计算得出的一个单元(例如:个人,教室,学校)被指配接受特定处理的概率。通过让处理组和对照组的协变量趋同,倾向性评分可用于减少选择偏差。
 +
 +
假设有一个二值处理标识Z,一个响应变量r,以及被观测的背景协变量X。倾向性评分定义为在给定背景变量条件下,单元接受处理的条件概率:
    
:<math>e(x) \ \stackrel{\mathrm{def}}{=}\  \Pr(Z=1 \mid X=x).</math>
 
:<math>e(x) \ \stackrel{\mathrm{def}}{=}\  \Pr(Z=1 \mid X=x).</math>
    
In the context of [[causal inference]] and [[survey methodology]], propensity scores are estimated (via methods such as [[logistic regression]], [[random forests]], or others), using some set of covariates. These propensity scores are then used as estimators for weights to be used with [[Inverse probability weighting]] methods.
 
In the context of [[causal inference]] and [[survey methodology]], propensity scores are estimated (via methods such as [[logistic regression]], [[random forests]], or others), using some set of covariates. These propensity scores are then used as estimators for weights to be used with [[Inverse probability weighting]] methods.
 +
 +
In the context of causal inference and survey methodology, propensity scores are estimated (via methods such as logistic regression, random forests, or others), using some set of covariates. These propensity scores are then used as estimators for weights to be used with Inverse probability weighting methods.
 +
 +
在因果推断和调查方法的范围内,使用一些协变量估计倾向性评分(通过逻辑回归、随机森林或其他方法)。然后将这些倾向性评分作为权重的估计量用于逆概率加权方法。
    
===Main theorems===
 
===Main theorems===
 +
 +
===主要定理===
 +
 
The following were first presented, and proven, by Rosenbaum and Rubin in 1983:<ref name="Rosenbaum 1983 41–55"/>
 
The following were first presented, and proven, by Rosenbaum and Rubin in 1983:<ref name="Rosenbaum 1983 41–55"/>
 +
 +
The following were first presented, and proven, by Rosenbaum and Rubin in 1983:[1]
 +
 +
以下是Rosenbaum和Rubin于1983年首次提出并证明的:[1]
    
* The propensity score <math>e(x)</math> is a balancing score.
 
* The propensity score <math>e(x)</math> is a balancing score.
 +
* 倾向性评分<math>e(x)</math>是平衡得分。
 +
 
* Any score that is 'finer' than the propensity score is a balancing score (i.e.: <math>e(X)=f(b(X))</math> for some function ''f''). The propensity score is the coarsest balancing score function, as it takes a (possibly) multidimensional object (''X''<sub>''i''</sub>) and transforms it into one dimension (although others, obviously, also exist), while <math>b(X)=X</math> is the finest one.
 
* Any score that is 'finer' than the propensity score is a balancing score (i.e.: <math>e(X)=f(b(X))</math> for some function ''f''). The propensity score is the coarsest balancing score function, as it takes a (possibly) multidimensional object (''X''<sub>''i''</sub>) and transforms it into one dimension (although others, obviously, also exist), while <math>b(X)=X</math> is the finest one.
 +
* 任何比倾向性评分“精细”的得分都是平衡得分(即:对于函数''f'',<math>e(X)=f(b(X))</math>)。倾向性评分是最粗粒度的平衡得分函数,因为它把一个(可能是)多维的对象(''X''<sub>''i''</sub>)转换成一维(尽管其他维度显然也存在),而<math>b(X)=X</math>则是最细粒度的平衡得分函数。
 +
 
* If treatment assignment is strongly ignorable given ''X'' then:
 
* If treatment assignment is strongly ignorable given ''X'' then:
 +
* 如果对于给定的''X'',处理分配满足强可忽略条件,则:
 +
 
:* It is also strongly ignorable given any balancing function. Specifically, given the propensity score:
 
:* It is also strongly ignorable given any balancing function. Specifically, given the propensity score:
 +
:* 给定任何的平衡函数,具体来说,给定倾向性评分,处理分配也是强可忽略的:
 +
 
:::<math> (r_0, r_1) \perp Z \mid e(X).</math>
 
:::<math> (r_0, r_1) \perp Z \mid e(X).</math>
 +
 
:* For any value of a balancing score, the difference between the treatment and control means of the samples at hand (i.e.: <math>\bar{r}_1-\bar{r}_0</math>), based on subjects that have the same value of the balancing score, can serve as an [[Bias of an estimator|unbiased estimator]] of the [[average treatment effect]]: <math>E[r_1]-E[r_0]</math>.  
 
:* For any value of a balancing score, the difference between the treatment and control means of the samples at hand (i.e.: <math>\bar{r}_1-\bar{r}_0</math>), based on subjects that have the same value of the balancing score, can serve as an [[Bias of an estimator|unbiased estimator]] of the [[average treatment effect]]: <math>E[r_1]-E[r_0]</math>.  
 +
:* 对于有相同平衡得分值的处理样本和对照样本,它们响应变量均值的差值(即:<math>\bar{r}_1-\bar{r}_0</math>),可以作为[[average treatment effect|平均处理效应]]的[[Bias of an estimator|无偏估计量]]:<math>E[r_1]-E[r_0]</math>。
 +
 
* Using sample estimates of balancing scores can produce sample balance on&nbsp;''X''
 
* Using sample estimates of balancing scores can produce sample balance on&nbsp;''X''
 +
* 利用平衡得分的样本估计可产生在X上平衡的样本
    
===Relationship to sufficiency===
 
===Relationship to sufficiency===
 +
===与充分性的关系===
    
If we think of the value of ''Z'' as a [[Statistical parameter|parameter]] of the population that impacts the distribution of ''X'' then the balancing score serves as a [[Sufficient_statistic#Mathematical_definition|sufficient statistic]] for ''Z''. Furthermore, the above theorems indicate that the propensity score is a [[Sufficient_statistic#Minimal_sufficiency|minimal sufficient statistic]] if thinking of ''Z'' as a parameter of ''X''. Lastly, if treatment assignment ''Z'' is strongly ignorable given ''X'' then the propensity score is a [[Sufficient_statistic#Minimal_sufficiency|minimal sufficient statistic]] for the joint distribution of <math>(r_0, r_1)</math>.
 
If we think of the value of ''Z'' as a [[Statistical parameter|parameter]] of the population that impacts the distribution of ''X'' then the balancing score serves as a [[Sufficient_statistic#Mathematical_definition|sufficient statistic]] for ''Z''. Furthermore, the above theorems indicate that the propensity score is a [[Sufficient_statistic#Minimal_sufficiency|minimal sufficient statistic]] if thinking of ''Z'' as a parameter of ''X''. Lastly, if treatment assignment ''Z'' is strongly ignorable given ''X'' then the propensity score is a [[Sufficient_statistic#Minimal_sufficiency|minimal sufficient statistic]] for the joint distribution of <math>(r_0, r_1)</math>.
 +
 +
If we think of the value of Z as a parameter of the population that impacts the distribution of X then the balancing score serves as a sufficient statistic for Z. Furthermore, the above theorems indicate that the propensity score is a minimal sufficient statistic if thinking of Z as a parameter of X. Lastly, if treatment assignment Z is strongly ignorable given X then the propensity score is a minimal sufficient statistic for the joint distribution of {\displaystyle (r_{0},r_{1})}{\displaystyle (r_{0},r_{1})}.
 +
 +
如果我们把''Z''的值想成影响''X''分布的群体参数,则平衡得分充当了''Z''的充分统计量。进一步,上述定理指出,如果把''Z''视为''X''的参数,则倾向性评分就是最小充分统计量。最后,给定''X'',如果''Z''是强可忽略的,则倾向性评分是<math>(r_0, r_1)</math>联合分布的最小统计量
 +
    
===Graphical test for detecting the presence of confounding variables===
 
===Graphical test for detecting the presence of confounding variables===
 +
===存在混杂变量的图检测方法===
    
[[Judea Pearl]] has shown that there exists a simple graphical test, called the back-door criterion, which detects the presence of confounding variables. To estimate the effect of treatment, the background variables X must block all back-door paths in the graph. This blocking can be done either by adding the confounding variable as a control in regression, or by matching on the confounding variable.<ref name="pearl">{{cite book |last=Pearl |first=J. |year=2000 |title=Causality: Models, Reasoning, and Inference |url=https://archive.org/details/causalitymodelsr0000pear |url-access=registration |location=New York |publisher=Cambridge University Press |isbn=978-0-521-77362-1 }}</ref>
 
[[Judea Pearl]] has shown that there exists a simple graphical test, called the back-door criterion, which detects the presence of confounding variables. To estimate the effect of treatment, the background variables X must block all back-door paths in the graph. This blocking can be done either by adding the confounding variable as a control in regression, or by matching on the confounding variable.<ref name="pearl">{{cite book |last=Pearl |first=J. |year=2000 |title=Causality: Models, Reasoning, and Inference |url=https://archive.org/details/causalitymodelsr0000pear |url-access=registration |location=New York |publisher=Cambridge University Press |isbn=978-0-521-77362-1 }}</ref>
    +
Judea Pearl has shown that there exists a simple graphical test, called the back-door criterion, which detects the presence of confounding variables. To estimate the effect of treatment, the background variables X must block all back-door paths in the graph. This blocking can be done either by adding the confounding variable as a control in regression, or by matching on the confounding variable.[2]
 +
 +
朱迪亚·珀尔Judea Pearl已经表明存在一个简单的图检测方法,称为后门准则,它可以检测到混杂变量的存在。为了估计处理效果,背景变量X必须阻断图中的所有后门路径。可以通过在回归的控制变量中加入混杂变量,或者在混杂变量上进行匹配来实现阻断。
    
==Disadvantages==
 
==Disadvantages==
 +
==缺点==
    
PSM has been shown to increase model "imbalance, inefficiency, model dependence, and bias," which is not the case with most other matching methods. The insights behind the use of matching still hold but should be applied with other matching methods; propensity scores also have other productive uses in weighting and doubly robust estimation.
 
PSM has been shown to increase model "imbalance, inefficiency, model dependence, and bias," which is not the case with most other matching methods. The insights behind the use of matching still hold but should be applied with other matching methods; propensity scores also have other productive uses in weighting and doubly robust estimation.
第232行: 第284行:  
【机器翻译】PSM 已经被证明会增加模型的“不平衡性、低效率、模型依赖性和偏差”,这与大多数其他匹配方法不同。使用匹配的见解仍然有效,但应该与其他匹配方法一起应用; 倾向得分在加权和双重稳健估计方面也有其他有益的用途。
 
【机器翻译】PSM 已经被证明会增加模型的“不平衡性、低效率、模型依赖性和偏差”,这与大多数其他匹配方法不同。使用匹配的见解仍然有效,但应该与其他匹配方法一起应用; 倾向得分在加权和双重稳健估计方面也有其他有益的用途。
   −
PSM 已经被证明会增加模型的“不平衡性、低效率、模型依赖性和偏差”,这与大多数其他匹配方法不同。使用匹配的见解仍然有效,但应该与其他匹配方法一起应用; 倾向得分在加权和双重稳健估计方面也有其他有益的用途。
+
PSM 已经被证明会加剧模型的“不平衡性、低效率、模型依赖性和偏差”,这与大多数其他匹配方法不同。匹配方法背后的见解仍然成立,但应该与其他匹配方法一起应用;倾向得分在加权和双重稳健估计方面也有其他有益的用途。
       
Like other matching procedures, PSM estimates an average treatment effect from observational data. The key advantages of PSM were, at the time of its introduction, that by using a linear combination of covariates for a single score, it balances treatment and control groups on a large number of covariates without losing a large number of observations. If units in the treatment and control were balanced on a large number of covariates one at a time, large numbers of observations would be needed to overcome the "dimensionality problem" whereby the introduction of a new balancing covariate increases the minimum necessary number of observations in the sample geometrically.
 
Like other matching procedures, PSM estimates an average treatment effect from observational data. The key advantages of PSM were, at the time of its introduction, that by using a linear combination of covariates for a single score, it balances treatment and control groups on a large number of covariates without losing a large number of observations. If units in the treatment and control were balanced on a large number of covariates one at a time, large numbers of observations would be needed to overcome the "dimensionality problem" whereby the introduction of a new balancing covariate increases the minimum necessary number of observations in the sample geometrically.
   −
与其他匹配程序一样,PSM 从观测数据中估计平均处理效果。在引入 PSM 的时候,它的主要优点是,通过使用一个线性组合的协变量作为一个单一的评分,它平衡了治疗组和对照组在大量的协变量上,而不会失去大量的观察数据。如果处理和控制中的单元在大量的协变量上一次平衡,就需要大量的观测数据来克服“维数问题”,即引入新的平衡协变量几何地增加样本中必要的最小观测数据。
+
【机器翻译】与其他匹配程序一样,PSM 从观测数据中估计平均处理效果。在引入 PSM 的时候,它的主要优点是,通过使用一个线性组合的协变量作为一个单一的评分,它平衡了治疗组和对照组在大量的协变量上,而不会失去大量的观察数据。如果处理和控制中的单元在大量的协变量上一次平衡,就需要大量的观测数据来克服“维数问题”,即引入新的平衡协变量几何地增加样本中必要的最小观测数据。
 +
 
 +
与其他匹配过程一样,PSM也是从观测数据中估计平均处理效应。在引入PSM之时,它的主要优点是,通过使用协变量的线性组合得到一个单一评分,以大量的协变量为基础平衡了处理组和对照组,却不大量损失观测数据。如果在有众多协变量的情况下,对每一个些变量都分别做处理单元和对照单元平衡的话,就需要大量的观测数据来克服”维度问题“,即,每引入一个新的平衡协变量都会在几何上增加最小所需的观测样本数量
       
One disadvantage of PSM is that it only accounts for observed (and observable) covariates and not latent characteristics. Factors that affect assignment to treatment and outcome but that cannot be observed cannot be accounted for in the matching procedure. As the procedure only controls for observed variables, any hidden bias due to latent variables may remain after matching. Another issue is that PSM requires large samples, with substantial overlap between treatment and control groups.
 
One disadvantage of PSM is that it only accounts for observed (and observable) covariates and not latent characteristics. Factors that affect assignment to treatment and outcome but that cannot be observed cannot be accounted for in the matching procedure. As the procedure only controls for observed variables, any hidden bias due to latent variables may remain after matching. Another issue is that PSM requires large samples, with substantial overlap between treatment and control groups.
   −
PSM 的一个缺点是它只能解释观察到的(和可观察到的)协变量,而不能解释潜在的特征。影响治疗分配和结果但无法观察的因素不能在匹配程序中说明。由于程序只控制观察变量,任何隐藏的偏见由于潜在变量可能仍然匹配后。另一个问题是 PSM 需要大量的样本,治疗组和对照组之间有大量的重叠。
+
【机器翻译】PSM 的一个缺点是它只能解释观察到的(和可观察到的)协变量,而不能解释潜在的特征。影响治疗分配和结果但无法观察的因素不能在匹配程序中说明。由于程序只控制观察变量,任何隐藏的偏见由于潜在变量可能仍然匹配后。另一个问题是 PSM 需要大量的样本,治疗组和对照组之间有大量的重叠。
    +
PSM的一个缺点是它只能概括已观测的(和可观测的)协变量,而无法概括潜在变量。那些能影响处理分配却不可观测的因素无法被纳入匹配过程的考虑范围。由于匹配过程只控制可观测变量,那些隐藏的偏倚在完成匹配后可能依然存在。另一个问题是PSM还要求在大量样本中,在处理组和对照组之间有大量的重叠。
       
General concerns with matching have also been raised by Judea Pearl, who has argued that hidden bias may actually increase because matching on observed variables may unleash bias due to dormant unobserved confounders. Similarly, Pearl has argued that bias reduction can only be assured (asymptotically) by modelling the qualitative causal relationships between treatment, outcome, observed and unobserved covariates. Confounding occurs when the experimenter is unable to control for alternative, non-causal explanations for an observed relationship between independent and dependent variables. Such control should satisfy the "backdoor criterion" of Pearl. It can also easily be implemented manually.
 
General concerns with matching have also been raised by Judea Pearl, who has argued that hidden bias may actually increase because matching on observed variables may unleash bias due to dormant unobserved confounders. Similarly, Pearl has argued that bias reduction can only be assured (asymptotically) by modelling the qualitative causal relationships between treatment, outcome, observed and unobserved covariates. Confounding occurs when the experimenter is unable to control for alternative, non-causal explanations for an observed relationship between independent and dependent variables. Such control should satisfy the "backdoor criterion" of Pearl. It can also easily be implemented manually.
   −
朱迪亚 · 珀尔也提出了关于配对的普遍担忧,他认为隐性偏见实际上可能会增加,因为观察变量的配对可能会由于潜在的未观察混杂因素而释放出偏见。同样,珀尔认为,只有通过建立治疗、结果、观察和未观察协变量之间的定性因果关系模型,才能确保(渐近地)减少偏见。当实验者无法控制对独立变量和因变量之间观察到的关系的替代性、非因果性解释时,混淆就发生了。这种控制应满足珍珠的“后门规范”。它也可以很容易地手动实现。
+
【机器翻译】朱迪亚 · 珀尔也提出了关于配对的普遍担忧,他认为隐性偏见实际上可能会增加,因为观察变量的配对可能会由于潜在的未观察混杂因素而释放出偏见。同样,珀尔认为,只有通过建立治疗、结果、观察和未观察协变量之间的定性因果关系模型,才能确保(渐近地)减少偏见。当实验者无法控制对独立变量和因变量之间观察到的关系的替代性、非因果性解释时,混淆就发生了。这种控制应满足珍珠的“后门规范”。它也可以很容易地手动实现。
 
  −
 
  −
Category:Regression analysis
  −
 
  −
类别: 回归分析
  −
 
  −
:* For any value of a balancing score, the difference between the treatment and control means of the samples at hand (i.e.: <math>\bar{r}_1-\bar{r}_0</math>), based on subjects that have the same value of the balancing score, can serve as an [[Bias of an estimator|unbiased estimator]] of the [[average treatment effect]]: <math>E[r_1]-E[r_0]</math>.
  −
 
  −
Category:Epidemiology
  −
 
  −
类别: 流行病学
  −
 
  −
* Using sample estimates of balancing scores can produce sample balance on&nbsp;''X''
  −
 
  −
Category:Observational study
  −
 
  −
类别: 观察性研究
  −
 
  −
 
     −
Category:Causal inference
+
Judea Pearl也提出了关于匹配方法的一般性担忧,他认为对可观测变量进行匹配可能会让那些原本处于休眠状态的混杂因素被释放,从而实际上可能加剧隐藏的偏倚。同样,Pearl认为,只有通过对处理、结果、可观测和不可观测的协变量之间的定性因果关系进行建模,才能确保(渐进地)减少偏倚。当试验者无法控制<font color="#32cd32">对独立变量和因变量之间观察到的关系的替代性、非因果性解释时</font>,混杂就会发生。这样的控制应该满足Pearl的“后门准则”。它很容易手工实现。
   −
类别: 因果推理
      
<noinclude>
 
<noinclude>
   −
<small>This page was moved from [[wikipedia:en:Propensity score matching]]. Its edit history can be viewed at [[倾向评分/edithistory]]</small></noinclude>
+
<small>This page was moved from [[wikipedia:en:Propensity score matching]]. Its edit history can be viewed at [[倾向性评分/edithistory]]</small></noinclude>
    
[[Category:待整理页面]]
 
[[Category:待整理页面]]
66

个编辑