调试代码后，似乎microsynth在处理tibble对象时存在一些问题;至少这是第二个错误的来源：函数newreshape尝试运行time.tmp <- data[,timevar]，当应用到tibble时，它不会提取向量（这会导致进一步向下的问题）--它应该是例如。time.tmp <- dplyr::pull(data, timevar)。
即使修复了这个问题，仍然有更多的错误，所以我会简单地尝试在运行代码之前将数据转换为data.frame，这对我很有效：

sea1 <- microsynth(as.data.frame(seattledmi), 
               idvar="ID", timevar="time", intvar="Intervention", 
               start.pre=1, end.pre=12, end.post=16, 
               match.out=match.out, match.covar=cov.var, 
               result.var=match.out, omnibus.var=match.out,
               test="lower",
               n.cores = min(parallel::detectCores(), 2))

这样有用吗

当数据中缺少观测值时，会发生此错误。下面的最小代码示例使用seattledmi数据集生成错误：

library(microsynth)

# Delete the fifth row of the seattledmi data set
mySeattledmi <- seattledmi[-c(5),]

# Define variables as in the microsynth tutorial
cov.var <- c("TotalPop", "BLACK", "HISPANIC", "Males_1521", "HOUSEHOLDS", 
             "FAMILYHOUS", "FEMALE_HOU", "RENTER_HOU", "VACANT_HOU")
match.out <- c("i_felony", "i_misdemea", "i_drugs", "any_crime")

# Run microsynth using the mySeattledmi data and all other parameters set equal to
# ... the values used in Example 1 of the microsynth tutorial
sea5 <- microsynth(mySeattledmi, 
                   idvar="ID", timevar="time", intvar="Intervention", 
                   start.pre=1, end.pre=12, end.post=16, 
                   match.out=match.out, match.covar=cov.var, 
                   result.var=match.out, omnibus.var=match.out,
                   test="lower",
                   n.cores = min(parallel::detectCores(), 2))

现在发生的事情是，microsynth的newresform函数正在创建一个矩阵，它以数据中的intvar列为值。该矩阵的维数为n乘t，其中n等于id的数量，t等于数据中的时间段的数量，例如：在Seattledmi数据中n = 9，642和t = 16。如果任何观测结果缺失，则该矩阵将包含NA。这导致if (dum <= dum1 + 1)中dum的值评估为NA和错误消息。
解决该问题的一种方法是消除任何包含缺失值的治疗组或对照组。继续前面的示例，下面的代码删除mySeattledmi数据中与小于16的ID（即end.post - start.pre + 1）观察结果。

library(dplyr)

nobsById <- mySeattledmi %>% group_by(ID) %>% summarize(nobs = n())
mySeattledmi2 <- mySeattledmi %>% subset(!(ID %in% nobsById[nobsById$nobs < 16, 1][[1]]))

# Call microsynth again this time with mySeattledmi2
sea6 <- microsynth(mySeattledmi2, 
                   idvar="ID", timevar="time", intvar="Intervention", 
                   start.pre=1, end.pre=12, end.post=16, 
                   match.out=match.out, match.covar=cov.var, 
                   result.var=match.out, omnibus.var=match.out,
                   test="lower",
                   n.cores = min(parallel::detectCores(), 2))