How can I fill missing values based on a condition in R

user_3227641 0
Viewed 714

I'm trying to fill some missing values for rows that meet certain conditions from another column. My data is given below. For China for the years 2002 and 2003, I want to copy values from the column named "manu_GDP_old" into the column "manu_GDP". In other words, I'm trying to fill the missing values of "manu_GDP" for China from the column "manu_GDP_old".

I would ideally like to do this using the dplyr package.

Thanks in advance.

df <- structure(list(country = c("Brazil", "Brazil", "Brazil", "Brazil", 
                           "Brazil", "China", "China", "China", "China", "China"), year = c(2002, 
                                                                                            2003, 2004, 2005, 2006, 2002, 2003, 2004, 2005, 2006), manu_GDP = c(12.3569626659174, 
                                                                                                                                                                14.4507645634139, 15.0995301566951, 14.7382811350657, 14.108945871671, 
                                                                                                                                                                NA, NA, 31.9750699702633, 32.0939243286777, 32.4523280565943), 
               manu_GDP_old = c(NA, NA, NA, NA, NA, "31.1", "32.5", "32.0", 
                                "32.1", "32.5")), row.names = c(NA, -10L), class = c("tbl_df", 
                                                                                     "tbl", "data.frame"))
dplyr conditional-statements r
2 Answers

dplyr Method

df %>%  
    mutate(manu_GDP = case_when(country == "China" & is.na(manu_GDP) ~ manu_GDP_old,
                                TRUE ~ as.character(manu_GDP)))

Base R Method

tf <- df$country == "China" & is.na(df$manu_GDP)
df$manu_GDP[tf] <- df$manu_GDP_old[tf]
user_161355200

One way would be:

df %>%
  mutate(manu_GDP = ifelse(is.na(manu_GDP), manu_GDP_old, manu_GDP))

or

df %>%
  mutate(manu_GDP = na_if(manu_GDP, manu_GDP_old))

in case you want to check at country level:

df %>%
  mutate(manu_GDP = ifelse(is.na(manu_GDP) & country == 'China',
                           manu_GDP_old,
                           manu_GDP))
user_73260580
Related