在R中使用full_join函数报错及缺失值补0的解决求助

1. 解决full_join报错问题 #

报错原因是你误将base R中merge()函数的all=TRUE参数用到了dplyr的full_join()里——full_join()本身就是全连接逻辑，不需要额外传递该参数，多余参数触发了报错。

把出错的代码行：

ESL <- full_join(ESL, ESL_SMBPN, all = TRUE, by = "Institution");

修改为：

ESL <- full_join(ESL, ESL_SMBPN, by = "Institution");

2. 拼接后缺失值替换为0 #

拼接完成后，用dplyr的across()函数批量将数值型列中的缺失值（NA）替换为0，不会影响非数值类型的列：

# 处理table1的缺失值
table1 <- table1 %>%
  mutate(across(where(is.numeric), ~replace(., is.na(.), 0)))

# 处理table1_Revised的缺失值
table1_Revised <- table1_Revised %>%
  mutate(across(where(is.numeric), ~replace(., is.na(.), 0)))

完整修正后的代码 #

library(dplyr)

# 执行全连接（移除错误的all=TRUE参数）
ESL <- full_join(ES_Liquidity, ES_Liquidity_prev, by = "Institution")
ESL_SMBPN <- full_join(ES_Liquidity_SMBPN, ES_Liquidity_SMBPN_prev, by = "Institution")
ESL <- full_join(ESL, ESL_SMBPN, by = "Institution")

table1 <- full_join(SC, ESL, by = "Institution")
table1_Revised <- full_join(SC_Revised, ESL, by = "Institution")

# 替换数值列的缺失值为0
table1 <- table1 %>%
  mutate(across(where(is.numeric), ~replace(., is.na(.), 0)))

table1_Revised <- table1_Revised %>%
  mutate(across(where(is.numeric), ~replace(., is.na(.), 0)))

内容的提问来源于stack exchange，提问作者aaliad