In this tutorial, we will learn how to create all possible combinations of two variables using tidyr’s expand_grid() function. For example, if we have two variables of interest and want to create a dataframe with all possible combinations of the values of the two variables, we can use expand_grid() function.
let us get started by loading tidyverse.
library(tidyverse)
Let us say we have two variables, each with 5 elements
var1 = letters[1:5] var1 ## [1] "a" "b" "c" "d" "e"
var2 = LETTERS[1:5] var2 ## [1] "A" "B" "C" "D" "E"
tidyr’s expand_grid() Example
And to create all possible combinations of these two variables, 25 combinations in total, we can use expand_grid() function in tidyr
combination_df <- expand_grid(var1= letters[1:5],
var2 = LETTERS[1:5])
combination_df
## # A tibble: 25 × 2
## var1 var2
## <chr> <chr>
## 1 a A
## 2 a B
## 3 a C
## 4 a D
## 5 a E
## 6 b A
## 7 b B
## 8 b C
## 9 b D
## 10 b E
## # ℹ 15 more rows
base R’s expand.grid() Example
tidyr’s expand_grid() function is inspired by base R’s expand.grid() function and it can create a dataframe with all possible combinations of factor variables as given in the example above.
combination_df <- expand.grid(var1= letters[1:5],
var2 = LETTERS[1:5])
combination_df
## var1 var2
## 1 a A
## 2 b A
## 3 c A
## 4 d A
## 5 e A
## 6 a B
## 7 b B
## 8 c B
## 9 d B
## 10 e B
Combinations of a dataframe and a variable wiht expand_grid()
One of the advantages of using tidyr’s expand_grid() function is that it can with dataframes and matrices, not just factor variables. Here is an example of using expand_grid() function in tidyr to create all possible combinations of a dataframe and a character/factor variable.
Here is a simple dataframe that we would like to expand for multiple companies.
df = tibble(year=2021, quarter=paste0("Q",1:4))
df
## # A tibble: 4 × 2
## year quarter
## <dbl> <chr>
## 1 2021 Q1
## 2 2021 Q2
## 3 2021 Q3
## 4 2021 Q4
With tidyr’s expand_grid(), we can create the combinations of dataframe and the factor/character variable.
expand_grid(df, companies = c("GOOG", "MSFT", "NVDA"))
## # A tibble: 12 × 3
## year quarter companies
## <dbl> <chr> <chr>
## 1 2021 Q1 GOOG
## 2 2021 Q1 MSFT
## 3 2021 Q1 NVDA
## 4 2021 Q2 GOOG
## 5 2021 Q2 MSFT
## 6 2021 Q2 NVDA
## 7 2021 Q3 GOOG
## 8 2021 Q3 MSFT
## 9 2021 Q3 NVDA
## 10 2021 Q4 GOOG
## 11 2021 Q4 MSFT
## 12 2021 Q4 NVDA