Reshape in R with variable name patterns

Question

I am trying to reproduce the results of a reshape in Stata using base R's reshape function.

Stata

webuse reshape3, clear
li, clean
// reshape long
reshape long inc@r ue, i(id) j(year)
list, sepby(id) clean

This produces, before the reshape:

. li, clean

       id   sex   inc80r   inc81r   inc82r   ue80   ue81   ue82  
  1.    1     0     5000     5500     6000      0      1      0  
  2.    2     1     2000     2200     3300      1      0      0  
  3.    3     0     3000     2000     1000      0      0      1

Note the pattern of the names for the stub inc. After the reshape, I get:

. list, sepby(id) clean

       id   year   sex   incr   ue  
  1.    1     80     0   5000    0  
  2.    1     81     0   5500    1  
  3.    1     82     0   6000    0  
  4.    2     80     1   2000    1  
  5.    2     81     1   2200    0  
  6.    2     82     1   3300    0  
  7.    3     80     0   3000    0  
  8.    3     81     0   2000    0  
  9.    3     82     0   1000    1

R

I run into trouble in R since I don't know how to specify the regular expressiokn required to parse the wide format variable names.

library(foreign)
dfReshape3 <- read.dta('http://www.stata-press.com/data/r12/reshape3.dta')
reshape(dfReshape3, dir='long', varying=3:8, v.names=c('inc', 'ue'),
        times = c('80', '81', '82'))

However, this gives me:

     id sex time  inc   ue
1.80  1   0   80 5000 5500
2.80  2   1   80 2000 2200
3.80  3   0   80 3000 2000
1.81  1   0   81 6000    0
2.81  2   1   81 3300    1
3.81  3   0   81 1000    0
1.82  1   0   82    1    0
2.82  2   1   82    0    0
3.82  3   0   82    0    1

Any help appreciated.

A really easy solution is to just remove the trailing "r" from the "inc[0-9]+r" variables. names(dfReshape3) <- gsub("r$", "", names(dfReshape3)). Then reshape is pretty straightforward: reshape(dfReshape3, dir='long', varying=3:8, sep = "") — A5C1D2H2I1M1N2O1R2T1

agstudy agstudy · Accepted Answer · 2013-02-03T13:51:56

You was really close, just give a list to varying

 reshape(dfReshape3, dir='long', varying=list(c(3:5),c(6:8)), v.names=c('inc', 'ue'),times = c('80', '81', '82'))
     id sex time  inc ue
1.80  1   0   80 5000  0
2.80  2   1   80 2000  1
3.80  3   0   80 3000  0
1.81  1   0   81 5500  1
2.81  2   1   81 2200  0
3.81  3   0   81 2000  0
1.82  1   0   82 6000  0
2.82  2   1   82 3300  0
3.82  3   0   82 1000  1

Reshape in R with variable name patterns

Stata

R

1 Answers