binning/looping/coding question

[Melanie Marten]

Hello!

I would like to create bins of a continous variable x. Each bin has a width of 0.05.

Below is the "non-loop" version of what I want to do:

capture drop bin
generate bin = .
replace bin =1 if x>=0 & x<0.05
replace bin =2 if x>=0.05 & x<0.1
replace bin =3 if x>=0.1 & x<0.15
....
replace bin =N if x>=1 & x<1.05

(therefore bin =. if x>=1.05)


It goes without saying that I would rather do the "loop" version. I came up with this code:

capture drop bin
generate bin =.
local k=1

forvalues i = 0(0.05)1 {
forvalues j = 0.05(0.05)1.05 {
replace bin = `k' if x >= `i' & x < `j'
local k = `k' + 1
}
}

The code appears to work, except for the naming of the bins: I don't have bins "1", "2" , "3"...N; instead I have bins "20", "40","60"...."400"

Can anyone help me fix this "issue"?

Many thanks!

[Ken Chui]

k advances 20 times inside the inner j-loop. Try move the counter to the outer i-loop:

forvalues i = 0(0.05)1 {
forvalues j = 0.05(0.05)1.05 {
replace bin = `k' if x >= `i' & x < `j'
}
local k = `k' + 1
}

[Nick Cox]

No need for any looping here. Your bin limits are all just multiples of 0.05. See this, compare the use of ceil() and note also

https://www.stata-journal.com/articl...article=dm0002

https://www.stata-journal.com/articl...article=dm0095

. clear

. set obs 1000
number of observations (_N) was 0, now 1,000

. set seed 3082

. gen y = runiform()

. gen bin = 0.05 * floor(y / 0.05)

. tab bin

        bin |      Freq.     Percent        Cum.
------------+-----------------------------------
          0 |         46        4.60        4.60
        .05 |         59        5.90       10.50
         .1 |         43        4.30       14.80
        .15 |         41        4.10       18.90
         .2 |         58        5.80       24.70
        .25 |         42        4.20       28.90
         .3 |         47        4.70       33.60
        .35 |         49        4.90       38.50
         .4 |         44        4.40       42.90
        .45 |         49        4.90       47.80
         .5 |         42        4.20       52.00
        .55 |         53        5.30       57.30
         .6 |         54        5.40       62.70
        .65 |         60        6.00       68.70
         .7 |         52        5.20       73.90
        .75 |         66        6.60       80.50
         .8 |         46        4.60       85.10
        .85 |         44        4.40       89.50
         .9 |         50        5.00       94.50
        .95 |         55        5.50      100.00
------------+-----------------------------------
      Total |      1,000      100.00

.

[Melanie Marten]

As always, many thanks for your clear response!