Create new column based on pattern in other column-CodePudding

I have the following data frame in R out of which I'd like to create a new column containing the Nut for each municipal (See second table). "Nut" refers simply to a higher hirachy level of municipalities in portugal. For later analysis I need to group the data by Nuts. The entire dataframe consists of 308 municipalities and 25 Nuts.

Does someone have a suggestion on how to approach this task? Since the number of municipals in each Nut differes I have difficulties on where to begin.

geo-group	nuts_municipal
Nut III	Alto Minho
Municipal	Arcos de Valdevez
Municipal	Caminha
Municipal	Monção
Municipal	Ponte da Barca
Nuts III	Ponte da Barca
Municipal	Amares
Municipal	Barcelos
Municipal	Braga
Nuts III	Fafe
Municipal	Ave

This is what I'd like to have as a final result.

geo-group	nuts_municipal	Nut
Nut III	Alto Minho
Municipal	Arcos de Valdevez	Alto Minho
Municipal	Caminha	Alto Minho
Municipal	Monção	Alto Minho
Municipal	Ponte da Barca	Alto Minho
Nut III	Cávado
Municipal	Amares	Cávado
Municipal	Barcelos	Cávado
Municipal	Braga	Cávado
Nut III	Ave
Municipal	Fafe	Ave
Municipal	Mondim de Basto	Ave

I have difficulties on where to begin and so far haven't found any appreach.

CodePudding user response：

A second option would be to use tidyr::fill and if_else:

library(tidyverse)

dat |> 
  mutate(Nut = if_else(grepl("^Nut", `geo-group`), nuts_municipal, NA_character_)) |> 
  tidyr::fill(Nut) |> 
  mutate(Nut = if_else(grepl("^Nut", `geo-group`), "", Nut))
#>    geo-group    nuts_municipal            Nut
#> 1    Nut III        Alto Minho               
#> 2  Municipal Arcos de Valdevez     Alto Minho
#> 3  Municipal           Caminha     Alto Minho
#> 4  Municipal            Monção     Alto Minho
#> 5  Municipal    Ponte da Barca     Alto Minho
#> 6   Nuts III    Ponte da Barca               
#> 7  Municipal            Amares Ponte da Barca
#> 8  Municipal          Barcelos Ponte da Barca
#> 9  Municipal             Braga Ponte da Barca
#> 10  Nuts III              Fafe               
#> 11 Municipal               Ave           Fafe

DATA

dat <- data.frame(
       check.names = FALSE,
       `geo-group` = c("Nut III","Municipal",
                       "Municipal","Municipal","Municipal","Nuts III","Municipal",
                       "Municipal","Municipal","Nuts III","Municipal"),
    nuts_municipal = c("Alto Minho",
                       "Arcos de Valdevez","Caminha","Monção","Ponte da Barca",
                       "Ponte da Barca","Amares","Barcelos","Braga","Fafe","Ave")
)

CodePudding user response：

You can group_by Nuts III and then use first:

df %>% 
  group_by(gp = cumsum(`geo-group` == "Nuts III")) %>% 
  mutate(Nut = ifelse(row_number() == 1, "", first(nuts_municipal)))