dividing observations into groups

anzas rustamaji pratama

Join Date: Nov 2016
Posts: 28

dividing observations into groups

04 Aug 2017, 06:07

dear STATA User
Apologies if this question was already asked, I'm having trouble in dividing my data. This is what I want to do:
1. I want to divide the variable "PERMNO" set into four equal groups, each group comprising a quarter of the data. The data is divided based on variable "total_assets" from the largest to smallest.
2. generate new variable, firm_size, with categorical 1,2,3 and 4, based on the four equal group.

"

Code:

* Example generated by -dataex-. To install: ssc install dataex clear input long(total_asset PERMNO) 1 69 2 70 3 441 4 95 5 90 6 464 7 109 8 98 9 224 10 108 11 274 12 65 13 67 14 528 15 45 16 319 17 424 18 284 19 113 20 4 21 93 22 13 23 281 24 499 25 315 26 323 27 25 28 161 29 199 30 290 31 368 32 163 33 188 34 261 35 506 36 37 37 482 38 326 39 311 40 207 41 478 42 237 43 80 44 198 45 485 46 295 47 280 48 401 49 279 50 423 51 362 52 486 53 96 54 400 55 56 56 38 57 136 58 430 59 240 60 197 61 531 62 180 63 23 64 535 65 239 66 371 67 260 68 283 69 480 70 71 71 491 72 75 73 135 74 422 75 292 76 57 77 357 78 426 79 440 80 431 81 417 82 329 83 179 84 55 85 350 86 361 87 351 88 508 89 182 90 86 91 208 92 287 93 174 94 175 95 40 96 231 97 205 98 369 99 509 100 169 end label values total_asset total_asset label def total_asset 1 "--", modify label def total_asset 2 "1003644400959490", modify label def total_asset 3 "101315624", modify label def total_asset 4 "101373000", modify label def total_asset 5 "102318460633088", modify label def total_asset 6 "1031213481984", modify label def total_asset 7 "1037602193408", modify label def total_asset 8 "1038706030936060", modify label def total_asset 9 "1044159856640", modify label def total_asset 10 "1050099200", modify label def total_asset 11 "1052131786752", modify label def total_asset 12 "105406005248000", modify label def total_asset 13 "10587951071232", modify label def total_asset 14 "105894649856", modify label def total_asset 15 "1063856111616", modify label def total_asset 16 "10683437547520", modify label def total_asset 17 "107160576", modify label def total_asset 18 "10733598277632", modify label def total_asset 19 "1075595264", modify label def total_asset 20 "1117684864", modify label def total_asset 21 "11359505874944", modify label def total_asset 22 "11377960812544", modify label def total_asset 23 "1142273081344", modify label def total_asset 24 "1142882435072", modify label def total_asset 25 "1150574788608", modify label def total_asset 26 "116375760", modify label def total_asset 27 "1175963435008", modify label def total_asset 28 "11840059867136", modify label def total_asset 29 "118806592", modify label def total_asset 30 "119437246464", modify label def total_asset 31 "119667792", modify label def total_asset 32 "1197796622336", modify label def total_asset 33 "1205850341376", modify label def total_asset 34 "1209792000", modify label def total_asset 35 "121215064", modify label def total_asset 36 "1214104469504", modify label def total_asset 37 "1218023129088", modify label def total_asset 38 "12257391017984", modify label def total_asset 39 "1228373622784", modify label def total_asset 40 "1229172506624", modify label def total_asset 41 "1229677568", modify label def total_asset 42 "12469009383424", modify label def total_asset 43 "12476256092160", modify label def total_asset 44 "1257609887744", modify label def total_asset 45 "12596824506368", modify label def total_asset 46 "1263115304960", modify label def total_asset 47 "1263726886912", modify label def total_asset 48 "1270812544", modify label def total_asset 49 "12779523145728", modify label def total_asset 50 "1288683913216", modify label def total_asset 51 "12922422034432", modify label def total_asset 52 "129799072", modify label def total_asset 53 "1299840106496", modify label def total_asset 54 "1300419969024", modify label def total_asset 55 "13057548877824", modify label def total_asset 56 "130701376", modify label def total_asset 57 "1314929508352", modify label def total_asset 58 "13185027072", modify label def total_asset 59 "1339032469504", modify label def total_asset 60 "1342118297600", modify label def total_asset 61 "1353634152448", modify label def total_asset 62 "1363641630720", modify label def total_asset 63 "136618852352", modify label def total_asset 64 "13734267322368", modify label def total_asset 65 "1381633359872", modify label def total_asset 66 "138196344111104", modify label def total_asset 67 "139701312", modify label def total_asset 68 "139809128448", modify label def total_asset 69 "14019294396416", modify label def total_asset 70 "140246496", modify label def total_asset 71 "1403237302272", modify label def total_asset 72 "14080570032128", modify label def total_asset 73 "1412285857792", modify label def total_asset 74 "1415136018432", modify label def total_asset 75 "14157428555776", modify label def total_asset 76 "14207414173696", modify label def total_asset 77 "1423242559488", modify label def total_asset 78 "14276198662144", modify label def total_asset 79 "1443540369408", modify label def total_asset 80 "1451906826240", modify label def total_asset 81 "145224416", modify label def total_asset 82 "14540108464128", modify label def total_asset 83 "14610863226880", modify label def total_asset 84 "14612273561600", modify label def total_asset 85 "14630080479232", modify label def total_asset 86 "147254256", modify label def total_asset 87 "14926224556032", modify label def total_asset 88 "149708177408", modify label def total_asset 89 "1504668416", modify label def total_asset 90 "1508215552", modify label def total_asset 91 "150879420416", modify label def total_asset 92 "15226009288704", modify label def total_asset 93 "1531365556224", modify label def total_asset 94 "1538089811968", modify label def total_asset 95 "1543216300032", modify label def total_asset 96 "1547569987584", modify label def total_asset 97 "1558285647872", modify label def total_asset 98 "1562589528064", modify label def total_asset 99 "1570230763520", modify label def total_asset 100 "1575467876352", modify label values PERMNO PERMNO label def PERMNO 4 "ABMM IJ Equity", modify label def PERMNO 13 "AGRO IJ Equity", modify label def PERMNO 23 "ALKA IJ Equity", modify label def PERMNO 25 "ALTO IJ Equity", modify label def PERMNO 37 "APOL IJ Equity", modify label def PERMNO 38 "ARGO IJ Equity", modify label def PERMNO 40 "ARNA IJ Equity", modify label def PERMNO 45 "ASDM IJ Equity", modify label def PERMNO 55 "AUTO IJ Equity", modify label def PERMNO 56 "BABP IJ Equity", modify label def PERMNO 57 "BACA IJ Equity", modify label def PERMNO 65 "BBKP IJ Equity", modify label def PERMNO 67 "BBMD IJ Equity", modify label def PERMNO 69 "BBNP IJ Equity", modify label def PERMNO 70 "BBRI IJ Equity", modify label def PERMNO 71 "BBRM IJ Equity", modify label def PERMNO 75 "BCIC IJ Equity", modify label def PERMNO 80 "BFIN IJ Equity", modify label def PERMNO 86 "BIPI IJ Equity", modify label def PERMNO 90 "BJBR IJ Equity", modify label def PERMNO 93 "BKSL IJ Equity", modify label def PERMNO 95 "BLTA IJ Equity", modify label def PERMNO 96 "BLTZ IJ Equity", modify label def PERMNO 98 "BMRI IJ Equity", modify label def PERMNO 108 "BORN IJ Equity", modify label def PERMNO 109 "BPFI IJ Equity", modify label def PERMNO 113 "BRMS IJ Equity", modify label def PERMNO 135 "CEKA IJ Equity", modify label def PERMNO 136 "CENT IJ Equity", modify label def PERMNO 161 "DILD IJ Equity", modify label def PERMNO 163 "DLTA IJ Equity", modify label def PERMNO 169 "DPUM IJ Equity", modify label def PERMNO 174 "DVLA IJ Equity", modify label def PERMNO 175 "DYAN IJ Equity", modify label def PERMNO 179 "ELTY IJ Equity", modify label def PERMNO 180 "EMDE IJ Equity", modify label def PERMNO 182 "ENRG IJ Equity", modify label def PERMNO 188 "ETWA IJ Equity", modify label def PERMNO 197 "GAMA IJ Equity", modify label def PERMNO 198 "GDST IJ Equity", modify label def PERMNO 199 "GDYR IJ Equity", modify label def PERMNO 205 "GLOB IJ Equity", modify label def PERMNO 207 "GMTD IJ Equity", modify label def PERMNO 208 "GOLD IJ Equity", modify label def PERMNO 224 "HOTL IJ Equity", modify label def PERMNO 231 "IDPR IJ Equity", modify label def PERMNO 237 "IMJS IJ Equity", modify label def PERMNO 239 "INAF IJ Equity", modify label def PERMNO 240 "INAI IJ Equity", modify label def PERMNO 260 "ITMA IJ Equity", modify label def PERMNO 261 "ITMG IJ Equity", modify label def PERMNO 274 "JTPE IJ Equity", modify label def PERMNO 279 "KBLV IJ Equity", modify label def PERMNO 280 "KBRI IJ Equity", modify label def PERMNO 281 "KDSI IJ Equity", modify label def PERMNO 283 "KICI IJ Equity", modify label def PERMNO 284 "KIJA IJ Equity", modify label def PERMNO 287 "KLBF IJ Equity", modify label def PERMNO 290 "KONI IJ Equity", modify label def PERMNO 292 "KPIG IJ Equity", modify label def PERMNO 295 "KREN IJ Equity", modify label def PERMNO 311 "LPPS IJ Equity", modify label def PERMNO 315 "MAGP IJ Equity", modify label def PERMNO 319 "MAPI IJ Equity", modify label def PERMNO 323 "MBAP IJ Equity", modify label def PERMNO 326 "MCOR IJ Equity", modify label def PERMNO 329 "MDLN IJ Equity", modify label def PERMNO 350 "MNCN IJ Equity", modify label def PERMNO 351 "MPMX IJ Equity", modify label def PERMNO 357 "MTFN IJ Equity", modify label def PERMNO 361 "MYOH IJ Equity", modify label def PERMNO 362 "MYOR IJ Equity", modify label def PERMNO 368 "NIKL IJ Equity", modify label def PERMNO 369 "NIPS IJ Equity", modify label def PERMNO 371 "NISP IJ Equity", modify label def PERMNO 400 "PORT IJ Equity", modify label def PERMNO 401 "POWR IJ Equity", modify label def PERMNO 417 "RAJA IJ Equity", modify label def PERMNO 422 "RELI IJ Equity", modify label def PERMNO 423 "RICY IJ Equity", modify label def PERMNO 424 "RIGS IJ Equity", modify label def PERMNO 426 "RMBA IJ Equity", modify label def PERMNO 430 "SAFE IJ Equity", modify label def PERMNO 431 "SAME IJ Equity", modify label def PERMNO 440 "SHID IJ Equity", modify label def PERMNO 441 "SHIP IJ Equity", modify label def PERMNO 464 "SONA IJ Equity", modify label def PERMNO 478 "SUGI IJ Equity", modify label def PERMNO 480 "SUPR IJ Equity", modify label def PERMNO 482 "TARA IJ Equity", modify label def PERMNO 485 "TBLA IJ Equity", modify label def PERMNO 486 "TBMS IJ Equity", modify label def PERMNO 491 "TIFA IJ Equity", modify label def PERMNO 499 "TMPI IJ Equity", modify label def PERMNO 506 "TPMA IJ Equity", modify label def PERMNO 508 "TRIL IJ Equity", modify label def PERMNO 509 "TRIM IJ Equity", modify label def PERMNO 528 "WAPO IJ Equity", modify label def PERMNO 531 "WIIM IJ Equity", modify label def PERMNO 535 "WSBP IJ Equity", modify

"

Thank you very much for you time,

Anzas

Tags: None

Nick Cox

Join Date: Mar 2014
Posts: 35458

04 Aug 2017, 08:18

In your example dataset there are no missing values and all values of total assets are distinct. So, this will work:

Code:

. sort total_asset

. gen group = ceil(4 * _n/_N)

. tab group

      group |      Freq.     Percent        Cum.
------------+-----------------------------------
          1 |         25       25.00       25.00
          2 |         25       25.00       50.00
          3 |         25       25.00       75.00
          4 |         25       25.00      100.00
------------+-----------------------------------
      Total |        100      100.00

. scatter total_asset group

but for a more general method check out fastxtile by Michael Stepner from SSC.

Comment

anzas rustamaji pratama

Join Date: Nov 2016

Posts: 28
#3

04 Aug 2017, 09:55

Thanks Nick Cox , it's solve my problem.
Comment
Italo Franz

Join Date: Oct 2017

Posts: 13
#4

04 May 2020, 09:53

Originally posted by Nick Cox View Post

In your example dataset there are no missing values and all values of total assets are distinct. So, this will work:

Code:

. sort total_asset . gen group = ceil(4 * _n/_N) . tab group group | Freq. Percent Cum. ------------+----------------------------------- 1 | 25 25.00 25.00 2 | 25 25.00 50.00 3 | 25 25.00 75.00 4 | 25 25.00 100.00 ------------+----------------------------------- Total | 100 100.00 . scatter total_asset group

but for a more general method check out fastxtile by Michael Stepner from SSC.

Hi Nick,

I found sometimes the ceil function is not that accurate, especially for large numbers of observations (e.g., > 30000 obs).

For example, you can take a look at the following codes:

Code:

clear use auto.dta * obs: 74 expand 4 * expand obs to 296 expand 296 * expand to 296 * 296 observations gen group = ceil(296 * _n/_N) tab group

You can see that some groups have 295, and some have 297.
Comment

Nick Cox

Join Date: Mar 2014
Posts: 35458

04 May 2020, 16:49

#4 is a salutary example. But don't blame ceil(): it works with what it is given. Every now and again what is given it is a smidgen out and (the problem) ceil() rounds the wrong way.

This is a way to explore the problem, which is our friend precision again.

In every observation that has observation number 296(296)87616 the answer should be an exact integer, If the result is a little too low, ceil() rounds up as elsewhere intended, but if it is a little too high ceil() rounds up by almost 1 and gives the wrong answer.

Code:

. clear

. set obs `=296^2'
number of observations (_N) was 0, now 87,616

. gen double foo = 296 * _n/_N

. format foo %21x

. l foo if foo != round(foo, 1) & mod(_n, 296) == 0

       +-----------------------+
       |                   foo |
       |-----------------------|
11544. | +1.37fffffffffffX+005 |
12728. | +1.57fffffffffffX+005 |
23088. | +1.37fffffffffffX+006 |
25456. | +1.57fffffffffffX+006 |
29304. | +1.8c00000000001X+006 |
       |-----------------------|
31672. | +1.ac00000000001X+006 |
34040. | +1.cbfffffffffffX+006 |
36408. | +1.ebfffffffffffX+006 |
46176. | +1.37fffffffffffX+007 |
47656. | +1.4200000000001X+007 |
       |-----------------------|
50912. | +1.57fffffffffffX+007 |
52392. | +1.6200000000001X+007 |
57128. | +1.81fffffffffffX+007 |
58608. | +1.8c00000000001X+007 |
61864. | +1.a1fffffffffffX+007 |
       |-----------------------|
63344. | +1.ac00000000001X+007 |
68080. | +1.cbfffffffffffX+007 |
69560. | +1.d600000000001X+007 |
72816. | +1.ebfffffffffffX+007 |
74296. | +1.f600000000001X+007 |
       +-----------------------+

Announcement

dividing observations into groups

Comment

Comment

Comment

Comment