Notice: On March 31, it was **announced** that Statalist is moving from an email list to a **forum**. The old list will shut down on April 23, and its replacement, **statalist.org** is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
"Vassilopoulos Achilleas" <vassilopoulos.statalist@gmail.com> |

To |
<statalist@hsphsun2.harvard.edu> |

Subject |
st: RE: problem with commands |

Date |
Tue, 7 Sep 2010 18:53:26 +0300 |

I think this is what you're looking for : cd "YOUR_DIRECTORY" gen firstsic_sub=int(sic_sub/100) gen firstsic_moth=int(sic_mother/100) gen twosic_sub=int(sic_sub/10) gen twosic_moth=int(sic_mother/10) capture erase new.dta tempname new postfile `new' str10 id_mother year double diver using new levelsof year_inc_sub, local(levels) foreach y of local levels { count if (year_inc_sub<=`y') & (firstsic_moth!=firstsic_sub) scalar digit=r(N) count if year_inc_sub<=`y' scalar total=r(N) scalar a=scalar(digit)*3/scalar(total) count if (year_inc_sub<=`y') & (firstsic_moth==firstsic_sub) scalar digit2=r(N) scalar b=scalar(digit2)*2/scalar(total) count if (year_inc_sub<=`y') & (twosic_moth==twosic_sub) scalar digit3=r(N) scalar c=scalar(digit3)*1/scalar(total) scalar all=scalar(a)+scalar(b)+scalar(c) post `new' ("X") (`y') (scalar(all)) scalar drop all digit total } postclose `new' use new, clear sort year Hope this helps, _____________ - _______________ Achilleas Vassilopoulos Agricultural University of Athens, Dept. of Agricultural Economics and Rural Development, Lab. of Political Economy and European Integration. Iera Odos 75, 11855, Athens, Greece Tel: (+30) 210-5294726 Fax: (+30) 2105294786 Email : avassilopoulos.aua@gmail.com -----Original Message----- From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of ???s???a G?a?????p????? Sent: 07 September, 2010 13:49 To: statalist@hsphsun2.harvard.edu Subject: st: problem with commands Dear Statalisters, I have a data set consisting of a number of companies which are all subsidiaries of the same company X. The following variables describe each of these companies and their mother company: 1) "id_mother", the identity of the mother company 2) "id_sub", the identity of the subsidiary company 3) "year_inc_sub", year of incorporation of the subsidiary 4) "SIC_mother", a 3-digit industry classification code for the mother 5) "SIC_sub", a 3-digit industry classification code for the subsidiary The dataset looks like this: id_mother sic_mother id_sub sic_sub year_inc_sub X 731 JP0JPN 738 1982 X 731 JP680JPN 737 2002 X 731 JP0JPN 899 2002 X 731 JP0JPN 355 1953 X 731 JPJPN 355 2001 X 731 JPN 502 1972 X 731 JPJPN 357 1960 Based on this information I want to create a new dataset with a time-series structure, which will refer to the mother company "X" for the period 1990-2005. In the new dataset I want to have the following variables: 1) "id_mother", as above; 2) "year", the year of the observation; 3) "diversification", a new variable whose estimation I explain below. The diversification variable should be created under the following logic: For observation in year 1990 the value of diversification should be the sum of a, b, and c below: a) take the total number of subsidiary companies which were incorporated in 1990 or earlier whose 1st digit of sic_sub is different than the 1st digit of sic_mother and multiply them by 3; divide this number by the total number of companies which where incorporated in 1990 or earlier (in the example, the total number of companies existing in 1990 is 4) b) take the total number of subsidiary companies which were incorporated in 1990 or earlier for which only the 1st digit of their sic_sub is the same as the 1st digit of sic_mother and multiply them by 2; divide this number by the total number of companies which where incorporated in 1990 or earlier c) take the total number of subsidiary companies which were incorporated in 1990 or earlier, whose 1st and 2nd digits of sic_sub are the same as the 1st and 2nd digits of sic_mother but they differ in the 3rd digit, and multiply them by 1, divide this number by the total number of companies which where incorporated in 1990 or earlier The same logic should apply for the observations for years 1991-2005 The expected dataset should look like this: id_mother year diversification X 1990 a+b+c when subsidiaries incorporated earlier than or equal to 1990 X 1991 a+b+c when subsidiaries incorporated earlier than or equal to 1991 X 1992 a+b+c when subsidiaries incorporated earlier than or equal to 1992 X ............................................................................ ................................ X 2005 a+b+c when subsidiaries incorporated earlier than or equal to 2005 - Hide quoted text - Thank you in advance. Chrysa * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/ * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/

**References**:**st: problem with commands***From:*Χρυσούλα Γιαννικοπούλου <chrygiann@gmail.com>

- Prev by Date:
**AW: st: Can you confirm these Stata limitations?** - Next by Date:
**st: RE: Counting Number of Week-Days** - Previous by thread:
**st: problem with commands** - Next by thread:
**st: RE: problem with commands** - Index(es):