Elimination of observations that do not vary over the period with R (r-cran)

Question

Good evening!

I have a base with the following characteristic:

Year   source   destination   HS04   value
1989    ARG        BRA        0101     1
1989    ARG        BRA        0102     0
1989    ARG        BRA        0103     0
1989    ARG        BRA        0104     1
  .      .          .           .      .
  .      .          .           .      .
  .      .          .           .      .
2010    ARG        BRA        0101     1
2010    ARG        BRA        0102     1
2010    ARG        BRA        0103     1
2010    ARG        BRA        0104     1

I need to eliminate the HS04 observations that did not vary over the period. That is, HS04 0101 and HS04 0104 since both initial and final years had a value of 1.

The reference would be HS04, ie for a given pair of countries (eg ARG and BRA) HS04 varied between the initial period and the final period.

The sample period covers 1989-2010

Thanks in advance for your attention!

score 2 · Answer 1 · answered Jun 29 '18 at 01:18

We can try

 data %>% group_by(HS04) %>% 
          mutate(flag = ifelse(min(value) & max(value)==1, 1, 0)) %>% 
          filter(flag==0) %>% ungroup()

Data

data <- read.table(text = "
               Year   source   destination   HS04   value
               1989    ARG        BRA        0101     1
               1989    ARG        BRA        0102     0
               1989    ARG        BRA        0103     0
               1989    ARG        BRA        0104     1
               2010    ARG        BRA        0101     1
               2010    ARG        BRA        0102     1
               2010    ARG        BRA        0103     1
               2010    ARG        BRA        0104     1
",header=T)

score 1 · Answer 2 · answered Jun 29 '18 at 05:03

1

using @A. Suliman data, n_distinct() exactly fits your need:

data %>% group_by(HS04) %>% 
      filter(n_distinct(value) >1) %>% ungroup()

answered Jun 29 '18 at 05:03

Xinlu

140
5

Hello, @A. Suliman and @Xinlu! The code is working perfectly, but I forgot to mention that it can only capture the change from zero to one. That is, the current code both captures the change from zero to one as well as from one to zero. However, this change from one to zero must also be eliminated. Can you help me? – Alexandre Loures Jun 30 '18 at 00:51

Elimination of observations that do not vary over the period with R (r-cran)

2 Answers2