R语言遍历数据集并存储线性模型系数及置信区间

用户拥有3个数据集（后续将扩展至1000个），希望对每个数据集运行相同的线性模型，存储模型系数及其置信上下限，并关联对应数据集名称。当前尝试的循环代码存在问题，无法正确获取所需信息，寻求高效实现方法。

原代码示例：

set.seed(1)
school1 = data.frame(student = sample(c(1:100), 100, r = T),
                     score = runif(100))
school2 = data.frame(student = sample(c(1:100), 100, r = T),
                     score = runif(100))
school3 = data.frame(student = sample(c(1:100), 100, r = T),
                     score = runif(100))
                     
schools = list('school1', 'school2', 'school3')
storage <- vector('list', length(schools))

for(i in seq_along(schools)){
  tmpdat <- schools[[i]]
  tmp <- lm(score ~ x1, data = tmpdat)
  storage[[i]] <- summary(tmp)$coef[1]
}

期望生成的结果格式：

WANT = data.frame(data = c('school1', 'school2', 'school3'),
                  coef = c(0,0,0),
                  coefLL = c(0,0,0),
                  coefUL=c(0,0,0))

原代码存在两个核心问题：

• schools列表存储的是数据集名称的字符串，而非实际的数据集对象，导致tmpdat是字符而非数据框，无法用于lm建模。
• 线性模型公式score ~ x1中的x1在数据集中不存在，实际可用的自变量是student。

针对大量数据集（1000个）的场景，推荐两种实现方式：

方式一：Base R 循环（无需额外依赖） #

先将所有数据集存入列表，再通过循环遍历每个数据集，提取所需信息后整合成结果数据框：

set.seed(1)
# 生成示例数据集
school1 = data.frame(student = sample(c(1:100), 100, r = T),
                     score = runif(100))
school2 = data.frame(student = sample(c(1:100), 100, r = T),
                     score = runif(100))
school3 = data.frame(student = sample(c(1:100), 100, r = T),
                     score = runif(100))

# 将数据集存入列表（关键：存对象而非字符串）
schools <- list(school1 = school1, school2 = school2, school3 = school3)

# 初始化结果存储
result <- data.frame(
  data = names(schools),
  coef = numeric(length(schools)),
  coefLL = numeric(length(schools)),
  coefUL = numeric(length(schools))
)

# 循环处理每个数据集
for (i in seq_along(schools)) {
  # 拟合线性模型
  model <- lm(score ~ student, data = schools[[i]])
  # 提取student的系数估计值（若要截距则取[[1]]）
  result$coef[i] <- coef(model)[[2]]
  # 计算95%置信区间
  ci <- confint(model, "student", level = 0.95)
  result$coefLL[i] <- ci[1]
  result$coefUL[i] <- ci[2]
}

# 查看结果
print(result)

方式二：使用purrr包（更简洁高效，适合大规模数据） #

purrr的map_dfr函数可以直接遍历列表并返回数据框，代码更简洁：

library(purrr)

set.seed(1)
school1 = data.frame(student = sample(c(1:100), 100, r = T),
                     score = runif(100))
school2 = data.frame(student = sample(c(1:100), 100, r = T),
                     score = runif(100))
school3 = data.frame(student = sample(c(1:100), 100, r = T),
                     score = runif(100))

schools <- list(school1 = school1, school2 = school2, school3 = school3)

# 定义处理单个数据集的函数
process_data <- function(data, name) {
  model <- lm(score ~ student, data = data)
  coef_val <- coef(model)[[2]]
  ci <- confint(model, "student", level = 0.95)
  data.frame(
    data = name,
    coef = coef_val,
    coefLL = ci[1],
    coefUL = ci[2]
  )
}

# 遍历所有数据集并合并结果
result <- imap_dfr(schools, process_data)

print(result)

• 若后续扩展到1000个数据集，建议通过list.files+read.csv等方式批量读取数据并存入列表，避免手动逐个定义数据集。
• 若需要提取截距的系数和置信区间，只需将代码中的"student"替换为"(Intercept)"，并调整coef(model)[[1]]即可。

内容的提问来源于stack exchange，提问作者bvowe