tmleMSM(tmle)
tmleMSM()所属R语言包：tmle
译者：生物统计家园网 机器人LoveR
描述----------Description----------
Targeted maximum likelihood estimation of the parameter of a marginal structural model (MSM) for binary point treatment effects. The tmleMSM function is minimally called with arguments (Y,A,W, MSM), where Y is a continuous or binary outcome variable, A is a binary treatment variable, (A=1 for treatment, A=0 for control), and W is a matrix or dataframe of baseline covariates. MSM is a valid regression formula for regressing Y on any combination of A, V, W, T, where V defines strata and T represents the time at which repeated measures on subjects are made. Missingness in the outcome is accounted for in the estimation procedure if missingness indicator Delta is 0 for some observations. Repeated measures can be identified using the id argument.
有针对性的最大似然估计的参数的边际结构模型（MSM）的二进制点的治疗效果。 tmleMSM，这里(Y,A,W, MSM)是连续或二进制结果变量，Y是一个二进制处理变量（A，A=1函数调用参数最小治疗，A=0控制），和W是一个矩阵或数据框的基线协变量。 MSM是一个有效的回归公式回归Y的任意组合A, V, W, T，其中V定义阶层和T的时间重复措施的主题。 Missingness的结果就是占估计过程中，如果missingness指标Delta为0的一些意见。 id使用参数，可以识别重复测量。
用法----------Usage----------
tmleMSM(Y, A, W, V, T = rep(1,length(Y)), Delta = rep(1, length(Y)), MSM,
v = NULL, Q = NULL, Qform = NULL, Qbounds = c(-Inf, Inf),
Q.SL.library = c("SL.glm", "SL.step", "SL.glm.interaction"),
cvQinit = FALSE, hAV = NULL, hAVform = NULL, g1W = NULL,
gform = NULL, pDelta1 = NULL, g.Deltaform = NULL,
g.SL.library = c("SL.glm", "SL.step", "SL.glm.interaction"),
ub = 1/0.025, family = "gaussian", fluctuation = "logistic",
alpha = 0.995, id = 1:length(Y), inference = TRUE, verbose = FALSE)
参数----------Arguments----------
参数：Y
continuous or binary outcome variable
连续或二进制结果变量
参数：A
binary treatment indicator, 1 - treatment, 0 - control
二进制处理指标，1 - 治疗，0 - 控制
参数：W
vector, matrix, or dataframe containing baseline covariates. Factors are not currently allowed.
包含基线协变量的向量，矩阵或数据框。目前尚未允许因素。
参数：V
vector, matrix, or dataframe of covariates used to define strata
用于确定地层的协变量的向量，矩阵或数据框
参数：T
optional time for repeated measures data
可选的时间重复测量数据
参数：Delta
indicator of missing outcome or treatment assignment. 1 - observed, 0 - missing
失踪的结果或处理分配指标。 1“ - 观察，0 - 失踪
参数：MSM
MSM of interest, specified as valid right hand side of a regression formula (see examples)
男男性接触者的利益，指定为有效右手边的回归公式（参见示例）
参数：v
optional value defining the strata of interest (V=v) for stratified estimation of MSM parameter
可选的价值为分层的MSM参数估计的定义各阶层的利益（V=v）
参数：Q
optional nx2 matrix of initial values for Q portion of the likelihood, (E(Y|A=0,W), E(Y|A=1,W))
可选的nx2矩阵的初始值Q部分的可能性，(E(Y|A=0,W), E(Y|A=1,W))
参数：Qform
optional regression formula for estimation of E(Y|A, W), suitable for call to glm
可选的回归公式估计E(Y|A, W)，适合呼叫glm的
参数：Qbounds
vector of upper and lower bounds on Y and predicted values for initial Q
矢量的Y值和预测值的上限和下限，用于初始Q
参数：Q.SL.library
optional vector of prediction algorithms to use for SuperLearner estimation of initial Q
可选的矢量预测算法使用SuperLearner估计的初始Q
参数：cvQinit
logical, if TRUE, estimates cross-validated predicted values using discrete super learning, default=FALSE
逻辑，如果TRUE，估计交叉验证的预测值离散超级学习，默认值=FALSE
参数：hAV
optional nx2 matrix used in numerator of weights for updating covariate and the influence curve. If unspecified, defaults to conditional probabilities P(A=1|V) or P(A=1|T), for repeated measures data. For unstabilized weights, pass in an nx2 matrix of all 1s
可选的nx2基质分子的权重更新协变量的影响曲线。如果未指定，则默认为条件概率P(A=1|V)或P(A=1|T)，重复测量数据。对于不稳定的权重，通过在nx2矩阵的所有1s
参数：hAVform
optionalregression formula of the form A~V+T, if specified this overrides the call to SuperLearner
optionalregression公式的形式A~V+T，如果指定覆盖调用SuperLearner
参数：g1W
optional vector of conditional treatment assingment probabilities, P(A=1|W)
可选的矢量有条件的的治疗assingment概率，P(A=1|W)
参数：gform
optional regression formula of the form A~W, if specified this overrides the call to SuperLearner
可选的回归公式的形式A~W，如果指定覆盖调用SuperLearner
参数：pDelta1
optional nx2 matrix of conditional probabilities for missingness mechanism,P(Delta=1|A=0,V,W,T), P(Delta=1|A=1,V,W,T).
可选的nx2missingness机制的条件概率矩阵，P(Delta=1|A=0,V,W,T), P(Delta=1|A=1,V,W,T)。
参数：g.Deltaform
optional regression formula of the form Delta~A+W, if specified this overrides the call to SuperLearner
可选的回归公式的形式Delta~A+W，如果指定覆盖调用SuperLearner
参数：g.SL.library
optional vector of prediction algorithms to use for SuperLearner estimation of g1W or pDelta1
可选的向量预测算法用于SuperLearnerg1W或pDelta1估计
参数：ub
upper bound on observation weights. See Details section for more information
观察权重的上界。 Details部分获取更多信息
参数：family
family specification for working regression models, generally "gaussian" for continuous outcomes (default), "binomial" for binary outcomes
系列规范工作回归模型，一般连续结果（默认）“高斯”，“二项式”二元结果
参数：fluctuation
"logistic" (default), or "linear"
“MF”（默认），或“线性”
参数：alpha
used to keep predicted initial values bounded away from (0,1) for logistic fluctuation
后勤波动用来保持界的距离预测的初始值（0,1）
参数：id
optional subject identifier
选修科目标识符
参数：inference
if TRUE, variance-covariance matrix, standard errors, pvalues, and 95% confidence intervals are calculated. Setting to FALSE saves a little time when bootstrapping.
如果TRUE，方差 - 协方差矩阵，标准的错误，pvalues，和95％的置信区间的计算。设置为false，在引导时节省一点时间。
参数：verbose
status messages printed if set to TRUE (default=FALSE)
状态消息时，本机设置为TRUE（默认值=FALSE）
Details
详细信息----------Details----------
ub bounds the IC by bounding the factor h(A,V)/[g(A,V,W)P(Delta=1|A,V,W)] between 0 and ub, default value = 1/0.025.
ub限定IC边界的因素h(A,V)/[g(A,V,W)P(Delta=1|A,V,W)]0ub，默认值= 1/0.025。
Q.SL.library Defaults to ("SL.glm", "SL.step", "SL.glm.interaction")
Q.SL.library默认值 - （SL.glm“，”SL.step“，”SL.glm.interaction）
g.SL.library Defaults to ("SL.glm", "SL.step", "SL.glm.interaction")
g.SL.library默认值 - （SL.glm“，”SL.step“，”SL.glm.interaction）
This choice is simply because these algorithms are included in the base R installation. See SuperLearner help files for further information.
此的选择是简单的，因为这些算法都包含在碱基R安装。请参阅SuperLearner的帮助文件的详细信息，。
值----------Value----------
参数：psi
MSM parameter estimate
MSM参数估计
参数：sigma
variance covariance matrix
方差协方差矩阵
参数：se
standard errors extracted from sigma
提取西格玛的标准误差
参数：pvalue
two-sided p-value
双面的p-值
参数：lb
lower bound on 95% confidence interval
在95％的置信区间的下界
参数：ub
upper bound on 95% confidence interval
在95％的置信区间的上限
参数：epsilon
fitted value of epsilon used to target initial Q
拟合值的ε目标初始Q
参数：psi.Qinit
MSM parameter estimate based on untargeted initial Q
基于MSM参数估计的不相关的初始Q
参数：Qstar
targeted estimate of Q, an nx2 matrix with predicted values for Q(0,W), Q(1,W) using the updated fit
有针对性的估计Q，nx2 Q(0,W), Q(1,W)使用更新的合适的预测值矩阵
参数：Qinit
initial estimate of Q. Qinit$coef are the coefficients for a glm model for Q, if applicable. Qinit$Q is an nx2 matrix, where n is the number of observations. Columns contain predicted values for Q(0,W),Q(1,W) using the initial fit. Qinit$type is method for estimating Q
初步估计的Q。 Qinit$coef是一个glm模型Q，如果适用的系数。 Qinit$Q是nx2矩阵，其中n的若干意见。列中包含的预测值Q(0,W),Q(1,W)使用初始的配合。 Qinit$type是方法估计Q
参数：g
treatment mechanism estimate. A list with three items: g$g1W contains estimates of P(A=1|W) for each observation, g$coef the coefficients for the model for g when glm used, g$type estimation procedure
治疗机理的估计。三个项目：g$g1W估计P(A=1|W)每个观察，g$coef系数的模型g当glm使用，<X列表>估计程序
参数：g.AV
estimate for h(A,V) or h(A,T). A list with three items: g.AV$g1W an nx2 matrix containing values of P(A=0|V,T), P(A=1|V,T) for each observation, g.AV$coef the coefficients for the model for g when glm used, g.AV$type estimation procedure
估计为h（A，V）或h（A，T）。三个项目：g.AV$g1Wnx2基质中含有值P(A=0|V,T), P(A=1|V,T)每个观察，g.AV$coef系数的模型g<X列表>使用，glm估计程序
参数：g_Delta
missingness mechanism estimate. A list with three items: g_Delta$g1W an nx2 matrix containing values of P(Delta=1|A,V,W,T) for each observation, g_Delta$coef the coefficients for the model for g when glm used, g_Delta$type estimation procedure
missingness机制的估计。三个项目：g_Delta$g1Wnx2基质中含有值P(Delta=1|A,V,W,T)每个观察，g_Delta$coef系数的模型g<X列表>使用，glm估计程序
（作者）----------Author(s)----------
Susan Gruber <a href="mailto:sgruber@cal.berkeley.edu">sgruber@cal.berkeley.edu</a>, in collaboration with Mark van der Laan.
参考文献----------References----------
参见----------See Also----------
summary.tmleMSM, estimateQ, estimateG, calcSigma, tmle
summary.tmleMSM，estimateQ，estimateG，calcSigma，tmle
实例----------Examples----------
library(tmle)
# Example 1. Estimating MSM parameter with correctly specified regression formulas[实施例1。估计MSM参数，正确指定的回归公式]
# MSM: psi0 + psi1*A + psi2*V + psi3*A*V (saturated)[MSM：Psi0成为+ PSI1 * A + PSI2 * V + psi3 * A * V（饱和）]
# true parameter value: psi = (0, 1, -2, 0.5) [真正的参数值：psi的=（0，-1，-2，0.5）]
# generate data[生成数据]
set.seed(100)
n <- 1000
W <- matrix(rnorm(n*3), ncol = 3)
colnames(W) <- c("W1", "W2", "W3")
V <- rbinom(n, 1, 0.5)
A <- rbinom(n, 1, 0.5)
Y <- rbinom(n, 1, plogis(A - 2*V + 0.5*A*V))
result.ex1 <- tmleMSM(Y, A, W, V, MSM = "A*V", Qform = Y~., gform = A~1,
hAVform = A~1, family = "binomial")
print(result.ex1)
# Example 2. Repeated measures data, two observations per id[实施例2。重复测量数据，观察每个ID]
# (e.g., crossover study design)[（例如，交叉研究，设计）]
# MSM: psi0 + psi1*A + psi2*V + psi3*V^2 + psi4*T[MSM：Psi0成为PSI1 * A + PSI2 * V + psi3 * V ^ 2 + psi4 * T]
# true parameter value: psi = (-2, 1, 0, 3, 0 )[真正的参数值：psi的=（-2，1，0，3，0）]
# generate data in wide format (id, W1, Y(t), W2(t), V(t), A(t)) [宽屏幕格式生成数据（身份证，Y（T），W1，W2（T），V（T），A（T））]
set.seed(100)
n <- 500
id <- rep(1:n)
W1 <- rbinom(n, 1, 0.5)
W2.1 <- rnorm(n)
W2.2 <- rnorm(n)
V.1 <- rnorm(n)
V.2 <- rnorm(n)
A.1 <- rbinom(n, 1, plogis(0.5 + 0.3 * W2.1))
A.2 <- 1-A.1
Y.1 <- -2 + A.1 - 2*V.1^2 + W2.1 + rnorm(n)
Y.2 <- -2 + A.2 - 2*V.2^2 + W2.2 + rnorm(n)
d <- data.frame(id, W1, W2=W2.1, W2.2, V=V.1, V.2, A=A.1, A.2, Y=Y.1, Y.2)
# change dataset from wide to long format[从广角到长格式的更改数据集]
longd <- reshape(d,
varying = cbind(c(3, 5, 7, 9), c(4, 6, 8, 10)),
idvar = "id",
direction = "long",
timevar = "T",
new.row.names = NULL,
sep = "")
# misspecified model for initial Q, partial misspecification for g[最初有Q误设模型，偏误设的g]
result.ex2 <- tmleMSM(Y = longd$Y, A = longd$A, W = longd[,c("W1", "W2")], V = longd$V,
T = longd$T, MSM = "A + V + I(V^2) + T", Qform = Y ~ A + V, gform = A ~ W2, id = longd$id)
print(result.ex2)
# Example 3: Introduce 20% missingness in example 2 data[例3：介绍20％missingness，例如2数据]
Delta <- rbinom(nrow(longd), 1, 0.8)
result.ex3 <- tmleMSM(Y = longd$Y, A = longd$A, W = longd[,c("W1", "W2")], V = longd$V, T=longd$T,
Delta = Delta, MSM = "A + V + I(V^2) + T", Qform = Y ~ A + V, gform = A ~ W2,
g.Deltaform = Delta~ 1, id=longd$id, verbose = TRUE)
print(result.ex3)
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 10:46:55
