使用map_dfr和summarize时如何处理含NA值的分组

当分组内目标指标全为NA时，lm()会因无有效样本报错，我们可以通过两种方式让程序返回NA而非崩溃，同时保留正确的分组结构：

方案1：提前检查有效样本量 #

先判断当前分组中目标指标的非NA值数量是否满足线性回归的最低要求（至少2个点），不满足直接返回NA：

library(dplyr)
library(tidyr)
library(purrr)

calculate_slope = function(df, measure) {
  df %>%
    summarize(
      Measure = measure,
      Slope = {
        # 提取当前分组的目标指标列
        target_vals = pull(., measure)
        # 非NA值不足2个时返回NA，否则拟合模型取斜率
        if(sum(!is.na(target_vals)) < 2) NA_real_ else {
          lm(reformulate("nDay", measure), data = .)$coefficients[2]
        }
      },
      .groups = "drop"
    )
}

# 测试代码
example_data = expand.grid(
                 Subject = 1:3,
                 Response = c("x", "y"),
                 nDay = 1:3
               ) %>%
               mutate(
                 A = runif(n(), 0, 1),
                 B = runif(n(), 0, 1),
                 C = runif(n(), 0, 1)
               ) %>%
               mutate(B = ifelse(Subject == 3, NA, B))

measures = c("A", "B", "C")
summary_data = map_dfr(measures, ~ example_data %>%
                         group_by(Subject, Response) %>%
                         calculate_slope(., .x)) %>%
               pivot_wider(names_from = Measure, values_from = Slope) %>%
               rename_with(~ paste0("slope_", .), -c(Subject, Response))

print(summary_data)

方案2：用tryCatch捕获拟合错误 #

如果想覆盖更多潜在报错场景（比如自变量nDay全相同），可以用tryCatch包裹lm()调用，捕获错误时返回NA：

calculate_slope = function(df, measure) {
  df %>%
    summarize(
      Measure = measure,
      Slope = tryCatch(
        lm(reformulate("nDay", measure), data = .)$coefficients[2],
        error = function(e) NA_real_
      ),
      .groups = "drop"
    )
}

两种方案的区别 #

• 方案1更高效，提前过滤无效分组，避免不必要的模型计算；
• 方案2更灵活，能处理所有导致lm()报错的情况，比如自变量无变异、因变量全NA等。

内容的提问来源于stack exchange，提问作者Zaida