mgm 包中的estimateNetwork 函数存在问题

我正在尝试使用 mgm 包运行混合图形模型。到目前为止，这是我的代码。我在最后一个命令周围加了星号，这是我似乎无法运行的命令。有人可以建议吗？

library(haven)
    > flordiss <- read_sav("C:/Users/schul/OneDrive/Desktop/flordiss.sav")
    > View(flordiss)
    > library(mgm)
    This is mgm 1.2-13
    Please report issues on Github: https://github.com/jmbh/mgm/issues
    > library(huge)
    > library(glasso)
    > library(qgraph)
    > library(bootnet)
    Loading required package: ggplot2
    This is bootnet 1.5
    For questions and issues, please see github.com/SachaEpskamp/bootnet.
    > mydata<-flordiss
    > mydata_matrix <- as.matrix(mydata)
    > my_model <- estimateNetwork(mydata_matrix, reg = "mgl", standardize = TRUE)
    Error in do.call(.input$estimator, c(list(data), .input$arguments)) : 
    'what' must be a function or character string
    In addition: Warning messages:
    1: In formals(fun) : argument is not a function
    2: In formals(fun) : argument is not a function
    > Corr_matrix <- cor(mydata)
    > imputed_data <- apply(numeric_data, 2, function(x) ifelse(is.na(x), mean(x, na.rm = TRUE), x))
    Error in apply(numeric_data, 2, function(x) ifelse(is.na(x), mean(x, na.rm = TRUE),  : 
    object 'numeric_data' not found
    > numeric_data <- mydata[, sapply(mydata, is.numeric)]
    > imputed_data <- apply(numeric_data, 2, function(x) ifelse(is.na(x), mean(x, na.rm = TRUE), x))
    > corr_matrix <- cor(imputed_data)
    > qgraph(corr_matrix, layout = "spring", labels = colnames(imputed_data))
    > **my_model <- estimateNetwork(corr_matrix, reg = "mgl", standardize = TRUE)
    Error in do.call(.input$estimator, c(list(data), .input$arguments)) : 
   'what' must be a function or character string
    In addition: Warning messages:
    1: In formals(fun) : argument is not a function
    2: In formals(fun) : argument is not a function
    > **

我重新加载了 mgm 和相关的软件包。我重新启动 R 并检查了之前的代码。

我建议您在 R 脚本中而不是在控制台中编写代码。例如，您现在还显示导致错误的代码行，然后显示不同的代码行来解决该错误，这使得阅读代码变得非常困难。在此，请注意mgm是一个函数mgm包装和estimateNetwork是一个函数bootnet作为包装器的包mgm如果你使用default = "mgm"。这在帮助文件中有解释?estimateNetwork。另请注意，输入estimateNetwork应该是原始数据而不是相关矩阵。