R（dplyr / tidyverse）| 使用mutate_at使用if_else语句构造一系列新变量

Question

I'm relatively new to this site and to the world of programming, so my apologies if this has already been asked.

Here's a modified version of a data frame I'm currently working with (truncated to make things easier to diagnose):

  COUNTRY          b_2010 c_2010 b_2011  c_2011   
1 Australia          50     62     67     56     
2 Austria            50     48     48     95      
3 Belgium            50     26     67     25      
4 Bulgaria           50     54     42     64      

Let's assume that I want to create a series of variables indicating that a country has a value equal to or greater than 50 for each existing variable in a given year .

I can do so by running something like this:

dataframe %>% mutate(d_2010 = if_else(b_2010 & c_2010 >= 50, "A", "B"),
                     d_2011 = if_else(b_2011 & c_2011 >= 50, "A", "B"))

This should produce the indicator variables I'm looking to construct, but the process will get awfully taxing if I have a lengthy time series. I'm sure there's a way to go about doing this more efficiently (using mutate_at or some other function), but I haven't been able to figure it out.

Can someone out there help me out?

Thanks!

Answer 1

In my eyes for " each existing variable in a given year " would be something like this:

dataframe %>% mutate(d_2010 = if_else(b_2010 >= 50 & c_2010 >= 50, "A", "B"),
                     d_2011 = if_else(b_2011 >= 50 & c_2011 >= 50, "A", "B"))

If this is the goal then I would calculate the new variables in a first step and join them later to the origin dataframe. Something like this:

df <- dataframe %>% 
  gather(starts_with("b_"), starts_with("c_"), key = Key , value = Value) %>% 
  mutate(Year = paste0("d_"str_sub(Key, 3, 6))) %>% # creat your new variable 
  group_by(COUNTRY, Year) %>% 
  summarise(d =  ifelse(prod(Value >=50), "A", "B")) %>% # calculate the value
  spread(Year, d)

# join both
dataframe <- dataframe %>% 
  left_join(df)