wfe(wfe)
wfe()所属R语言包：wfe
Fitting the Weighted Fixed Effects Model for Causal Inference
拟合加权固定效应模型的推论
译者：生物统计家园网 机器人LoveR
描述----------Description----------
wfe is used to fit weighted fixed effects model for causal inference. wfe also derives the regression weights for different causal quantity of interest.
wfe来拟合加权固定效应模型进行因果推理。 wfe也派生的回归权重不同的因果数量利益。
用法----------Usage----------
wfe(formula, data, treat = "treat.name",
unit.index, time.index = NULL, method = "unit",
qoi = c("ate", "att") , estimator = NULL, C.it = NULL,
hetero.se = TRUE, auto.se = TRUE,
White = TRUE, White.alpha = 0.05,
verbose = TRUE, unbiased.se = FALSE, unweighted = FALSE,
rank.check = FALSE, tol = sqrt(.Machine$double.eps))
参数----------Arguments----------
参数：formula
a symbolic description of the model to be fitted. The formula should not include dummmies for fixed effects. The details of model specifications are given under "Details".
安装一个象征性的模型来描述。公式不应该包括固定效应dummmies为。型号规格的详细信息刊载在“详细信息”。
参数：data
data frame containing the variables in the model.
数据框包含在模型中的变量。
参数：treat
a character string indicating the name of treatment variable used in the models. The treatment should be binary indicator (integer with 0 for the control group and 1 for the treatment group).
一个字符串，指示处理变量在模型中使用的名称。治疗指标应该是二进制的（整数为治疗组与对照组为0和1）。
参数：unit.index
a character string indicating the name of unit variable used in the models. The index of unit should be factor.
一个字符串，表示单位在模型中使用的变量的名称。单位应该是该指数的因素。
参数：time.index
a character string indicating the name of time variable used in the models. The index of time should be factor.
一个字符串表示模型所用的时间变量的名称。该指数的时间应的因素。
参数：method
method for weighted fixed effects regression, either unit for unit fixed effects; time for time fixed effects. The default is unit.
加权固定效应回归的方法，无论是unit为单位的固定效应，“time时间固定效应。默认的unit。
参数：qoi
one of "ate" or "att". The default is "ate".
一个"ate"或"att"。默认的"ate"。
参数：estimator
an optional character string indicating the estimating method. One of "fd" or "did". The default is NULL.
一个可选的字符串表示的估计方法。一个"fd"或"did"。默认的NULL。
参数：C.it
an optional non-negative numeric vector specifying relative weights for each unit of analysis.
一个可选的非负数值的相对权重向量确定各单位的分析。
参数：hetero.se
a logical value indicating whether heteroskedasticity across units is allowed in calculating standard errors. The default is TRUE.
一个逻辑值，该值指示是否允许异方差性跨部门的计算标准误差。默认的TRUE。
参数：auto.se
a logical value indicating whether arbitrary autocorrelation is allowed in calculating standard errors. The default is TRUE.
一个逻辑值，该值指示是否允许任意的自相关计算标准误差。默认的TRUE。
参数：White
a logical value indicating whether White misspecification statistics should be calculated. The default is TRUE.
一个逻辑值，该值指示是否应计算白误设统计。默认的TRUE。
参数：White.alpha
level of functional specification test. See White (1980) and Imai and Kim (2012). The default is 0.05.
功能规格测试的水平。请参阅白皮书（1980年）和今井和Kim（2012年）。默认的0.05。
参数：verbose
logical. If TRUE, helpful messages along with a progress report of the weight calculation are printed on the screen. The default is TRUE.
逻辑。如果TRUE，有用的信息以及权重计算的进度报告显示在屏幕上。默认的TRUE。
参数：unbiased.se
logical. If TRUE, bias-asjusted heteroskedasticity-robust standard errors are used. See Stock and Watson (2008). Should be used only for balanced panel. The default is FALSE.
逻辑。如果TRUE，偏置asjusted异方差稳健标准误差所使用。 Stock和Watson（2008年）。应仅用于平衡面板。默认的FALSE。
参数：unweighted
logical. If TRUE, standard unweighted fixed effects model is estimated. The default is FALSE.
逻辑。如果TRUE，标准不加权的固定效应模型估计。默认的FALSE。
参数：rank.check
logical. If TRUE, rank condition is checked for fast calculation with projection method. The default is FALSE.
逻辑。如果TRUE，秩条件检查投影方法的快速计算。默认的FALSE。
参数：tol
a relative tolerance to detect zero singular values for generalized inverse. The default is sqrt(.Machine$double.eps)
一个相对宽容的检测广义逆的奇异值为零。默认值是开方机（double.eps）
Details
详细信息----------Details----------
To fit the weighted unit (time) fixed effects model, use the syntax for the formula, y ~ x1 + x2, where y is a dependent variable and x1 and x2 are unit (time) varying covariates.
为了适应固定效应模型的加权单位（时间），使用的语法的公式，y ~ x1 + x2，y是一个因变量，x1和x2单位（时间）变化的协变量。
wfe calculates weights based on different underlying causal quantity of interest: Average Treatment Effect (qoi = "ate") or Average Treatment Effect for the Treated (qoi = "att").
wfe计算的权重根据不同的基础因果关系的关注量：一般治疗效果（qoi = "ate"）或平均治疗效果的治疗（qoi = "att"）。
One can further set estimating methods: First-Difference (estimator ="fd") or Difference-in-differences (estimator = "did"). For the two-way fixed effects model, set estimator = "did"
我们可以进一步设置估算方法：一阶差分（estimator ="fd"）或差的差异（estimator = "did"）。对于双向固定效应模型，设置estimator = "did"
To specify different ex-ante weights for each unit of analysis, use non-negative weights C.it. For instance, using the survey weights for C.it enables the estimation fo the average treatement effect for the target population.
要指定不同的权重事前各单位的分析，使用非负权重C.it。例如，使用的调查权重为C.it使估计为的平均treatement影响的目标人群。
An object of class "wfe" contains vectors of unique unit(time) names and unique unit(time) indices.
“WFE”类的一个对象包含向量的独特的单位（时间）的名称和独特的单位（时间）指数。
值----------Value----------
wfe returns an object of class "wfe", a list that contains the components listed below.
wfe返回一个对象类“WFE”，一个列表，包含以下列出的组件。
The function summary (i.e., summary.wfe) can be used to obtain a table of the results.
该函数summary（即，summary.wfe）可以用来获得的结果的表。
参数：coefficients
a named vector of coefficients
一个命名的系数向量
参数：residuals
the residuals, that is respons minus fitted values
残差，即反应作者减去拟合值
参数：df
the degree of freedom
的自由度
参数：W
weight matrix calculated from the model. Row and column indices can be found from unit.name, time.name.
模型计算出的权重矩阵。行和列索引，可以发现从unit.name，time.name。
参数：call
the matched call
匹配的呼叫
参数：causal
causal quantity of interest
因果数量利益
参数：estimator
the estimating method
估算方法
参数：unit.name
a vector containing unique unit names
一个向量，包含独特的单位名称
参数：unit.index
a vector containing unique unit index number
一个向量，包含独特的单位索引号
参数：time.name
a vector containing unique time names
一个向量，包含独特的时间名称
参数：time.index
a vector containing unique time index number
一个向量，包含独特的时间索引号
参数：method
call of the method used
要求所使用的方法
参数：vcov
the variance covariance matrix
方差协方差矩阵
参数：White.alpha
the alpha level for White specification test
白规格测试的alpha水平
参数：White.pvalue
the p-value for White specification test
白规格测试的p值
参数：White.stat
the White statistics
白统计
参数：x
the design matrix
设计矩阵
参数：y
the response vector
响应矢量
参数：mf
the model frame
模型框架
（作者）----------Author(s)----------
Kosuke Imai, Princeton University, <a href="mailto:kimai@princeton.edu">kimai@princeton.edu</a>
and In Song Kim, Princeton University, <a href="mailto:insong@princeton.edu">insong@princeton.edu</a>
参考文献----------References----------
Effects Regression Models for Causal Inference.” Technical Report, Department of Politics, Princeton University. available at http://imai.princeton.edu/research/FEmatch.html
Standard Errors for Fixed Effect Panel Data Regression” Econometrica, 76, 1.
Regression Functions.” International Economic Review, 21, 1, 149–170.
参见----------See Also----------
pwfe for fitting weighted fixed effects models with propensity score weighting
pwfe倾向评分加权拟合加权固定效应模型
实例----------Examples----------
### NOTE: this example illustrates the use of wfe function with randomly[＃＃注意：这个例子说明了随机的WFE功能与使用]
### generated panel data with arbitrary number of units and time.[＃＃生成任意数量的面板数据单位和时间。]
## generate panel data with number of units = N, number of time = Time[＃生成面板数据的单位数= N，时间=时间]
N <- 10 # number of distinct units[不同的单元的数量]
Time <- 15 # number of distinct time[不同的时间的数量]
## treatment effect[＃治疗效果]
beta <- 1
## generate treatment variable[＃生成处理变量]
treat <- matrix(rbinom(N*Time, size = 1, 0.25), ncol = N)
## make sure at least one observation is treated for each unit[＃确保每个单元至少一个观察治疗]
while ((sum(apply(treat, 2, mean) == 0) > 0) | (sum(apply(treat, 2, mean) == 1) > 0) |
(sum(apply(treat, 1, mean) == 0) > 0) | (sum(apply(treat, 1, mean) == 1) > 0)) {
treat <- matrix(rbinom(N*Time, size = 1, 0.25), ncol = N)
}
treat.vec <- c(treat)
## unit fixed effects[＃机组固定效应]
alphai <- rnorm(N, mean = apply(treat, 2, mean))
## geneate two random covariates[＃geneate两个随机协变量]
x1 <- matrix(rnorm(N*Time, 0.5,1), ncol=N)
x2 <- matrix(rbeta(N*Time, 5,1), ncol=N)
x1.vec <- c(x1)
x2.vec <- c(x2)
## generate outcome variable[＃产生的结果变量]
y <- matrix(NA, ncol = N, nrow = Time)
for (i in 1:N) {
y[, i] <- alphai[i] + treat[, i] + x1[,i] + x2[,i] + rnorm(Time)
}
y.vec <- c(y)
## generate unit and time index[＃生成单元和时间指数]
unit.index <- rep(1:N, each = Time)
time.index <- rep(1:Time, N)
Data.str <- as.data.frame(cbind(y.vec, treat.vec, unit.index, x1.vec, x2.vec))
colnames(Data.str) <- c("y", "tr", "strata.id", "x1", "x2")
Data.obs <- as.data.frame(cbind(y.vec, treat.vec, unit.index, time.index, x1.vec, x2.vec))
colnames(Data.obs) <- c("y", "tr", "unit", "time", "x1", "x2")
############################################################[＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃ ＃＃＃＃＃＃＃＃＃]
# Example 1: Stratified Randomized Experiments[例1：采用分层随机试验]
############################################################[＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃ ＃＃＃＃＃＃＃＃＃]
## run the weighted fixed effect regression with strata fixed effect.[＃运行的加权固定效应回归与地层固定作用。]
## Note: the quantity of interest is Average Treatment Effect ("ate")[注：关注量平均处理效果（“吃”）]
## and the standard errors allow heteroskedasticity and arbitrary[＃和标准误差允许异方差和任意]
## autocorrelation.[＃自相关。]
### Average Treatment Effect[＃＃平均处理效果]
mod.ate <- wfe(y~ tr+x1+x2, data = Data.str, treat = "tr",
unit.index = "strata.id", method = "unit",
qoi = "ate", hetero.se=TRUE, auto.se=TRUE)
## summarize the results[＃总结的结果。]
summary(mod.ate)
### Average Treatment Effect for the Treated[＃＃平均处理后的治疗效果]
mod.att <- wfe(y~ tr+x1+x2, data = Data.str, treat = "tr",
unit.index = "strata.id", method = "unit",
qoi = "att", hetero.se=TRUE, auto.se=TRUE)
## summarize the results[＃总结的结果。]
summary(mod.att)
############################################################[＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃ ＃＃＃＃＃＃＃＃＃]
# Example 2: Observational Studies with Unit Fixed-effects[例2：单位固定效应的观测研究]
############################################################[＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃ ＃＃＃＃＃＃＃＃＃]
## run the weighted fixed effect regression with unit fixed effect.[＃执行单位固定效应的加权固定效应回归。]
## Note: the quantity of interest is Average Treatment Effect ("ate")[注：关注量平均处理效果（“吃”）]
## and the standard errors allow heteroskedasticity and arbitrary[＃和标准误差允许异方差和任意]
## autocorrelation.[＃自相关。]
mod.obs <- wfe(y~ tr+x1+x2, data = Data.obs, treat = "tr",
unit.index = "unit", time.index = "time", method = "unit",
qoi = "ate", hetero.se=TRUE, auto.se=TRUE,
White = TRUE, White.alpha = 0.05)
## summarize the results[＃总结的结果。]
summary(mod.obs)
## extracting weigths[＃提取weigths]
summary(mod.obs)$W
###################################################################[＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃ ＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃]
# Example 3: Observational Studies with differences-in-differences[例3：观察性研究的差异，在差异]
###################################################################[＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃ ＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃]
## run difference-in-differences estimator.[＃运行差异，差异的估计。]
## Note: the quantity of interest is Average Treatment Effect ("ate")[注：关注量平均处理效果（“吃”）]
## and the standard errors allow heteroskedasticity and arbitrary[＃和标准误差允许异方差和任意]
## autocorrelation.[＃自相关。]
mod.did <- wfe(y~ tr+x1+x2, data = Data.obs, treat = "tr",
unit.index = "unit", time.index = "time", method = "unit",
qoi = "ate", estimator ="did", hetero.se=TRUE, auto.se=TRUE,
White = TRUE, White.alpha = 0.05, verbose = TRUE)
## summarize the results[＃总结的结果。]
summary(mod.did)
## extracting weigths[＃提取weigths]
summary(mod.did)$W
转载请注明:出自 生物统计家园网(http://www.biostatistic.net)。
注：
注1：为了方便大家学习，本文档为生物统计家园网机器人LoveR翻译而成，仅供个人R语言学习参考使用，生物统计家园保留版权。
注2：由于是机器人自动翻译，难免有不准确之处，使用时仔细对照中、英文内容进行反复理解，可以帮助R语言的学习。
注3：如遇到不准确之处，请在本贴的后面进行回帖，我们会逐渐进行修订。
|
发表于 2012-10-1 21:03:47
