Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

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

From |
erik.aadland@bi.no |

To |
statalist@hsphsun2.harvard.edu |

Subject |
st: st: Re: st: Re: st: generating variables for group members conditional on member's own values and values for other group members |

Date |
Tue, 11 Jan 2011 11:05:20 +0100 |

Thanks again, Eric. Your input really helped med along and I am grateful. Regarding your suggestion to reshape to wide at some points during the analysis, did you have any specific issues in mind? Sincerely, Erik Aadland. -----Forwarded by Erik Aadland/people/BISTIFT on 01/11/2011 10:59AM ----- To: <statalist@hsphsun2.harvard.edu> From: Eric Booth <ebooth@ppri.tamu.edu> Sent by: owner-statalist@hsphsun2.harvard.edu Date: 01/10/2011 05:07PM Subject: st: Re: st: Re: st: generating variables for group members conditional on member's own values and values for other group members‏ <> I think this is what you are asking for: ***added to end of prev example--> bys project_id: egen x2_new = max(x2) replace x2_new = . if x2==org_new *** The distinction of orgs i & j for questions like these becomes muddled because as you loop over the observations, every org is at some point i or j within each project_id. During your analysis, you may find it useful to, at some points, reshape your data to wide to help work across orgs in a project. - Eric __ Eric A. Booth Public Policy Research Institute Texas A&M University ebooth@ppri.tamu.edu On Jan 10, 2011, at 3:47 AM, <erik.aadland@bi.no> <erik.aadland@bi.no> wrote: > Dear Eric and statalist. > > Thank you for helping me along. > When I run the suggested code it generates the desired values for org_id (j) rather than for org_id (i). > > Is there a way to generate the desired values for org_id (i)? > > Sincerely, > > Erik Aadland. > > > <> > > Assuming there are no cases in your full dataset where there are more > than one case in each project with a new niche value==1 & > has nonmissing org_elite value) : > > ***********************! > /* setup */ > clear > inp project_id org_id org_elite org_new new_niche > 1 1 1 0.9 0 > 1 2 0.9 0.2 1 > 1 3 0.1 0.3 0 > 1 6 0 0 0 > 2 4 0.5 0.86 1 > 2 5 . . 0 > 2 1 1 0.9 0 > 3 6 0.5 0.9 1 > 3 7 . 0.9 0 > end > > /* QUESTION 1 > I am trying to generate a variable X by project_id > and org_id that does the following: > if org_id (i) has an org_elite value that is ( > 0 <.) > and collaborates with another org_id (j) > with an org_new value (> 0.85) and that org_id (j) > has a new niche value =1, then org_id (i) gets a > value of 1 and 0 otherwise on the new variable X. > */ > > > bys project_id: egen m_org_new = max(org_new) /// > if new_niche==1 > g x = 1 if m_org_new > .85 & /// > org_elite>0 & !mi(org_elite, m_org_new) > recode x (.=0) > > > > /* QUESTION 2 > Similarly, I am trying to generate a variable X2 by > project_id and org_id that does the following: > if org_id (i) has a new niche value =1, and collaborates > with another org_id (j) that has an org_elite value ( <.), > then org_id (i) gets a value on the new variable X2 that > equals the org_elite value for org_id (j). > */ > > g x2 = m_org_new if !mi(m_org_new, org_elite) > > ************************! > > - Eric > > __ > Eric A. Booth > Public Policy Research Institute > Texas A&M University > ebooth@ppri.tamu.edu > Office: +979.845.6754 > Fax: +979.845.0249 > http://ppri.tamu.edu > > > On Jan 9, 2011, at 11:36 AM, <erik.aadland@bi.no> > wrote: > >> project_id org_id org_elite org_new new_niche >> 1 1 1 0.9 0 >> 1 2 0.9 0.2 1 >> 1 3 0.1 0.3 0 >> 1 6 0 0 0 >> 2 4 0.5 0.86 1 >> 2 5 . . 0 >> 2 1 1 0.9 0 >> 3 6 0.5 0.9 1 >> 3 7 . 0.9 0 >> >> I am trying to generate a variable X by project_id and org_id that does the following: >> >> if org_id (i) has an org_elite value that is ( > 0 <.) , and collaborates with another org_id (j) with an org_new value (> 0.85) and that org_id (j) has a new niche value =1, then org_id (i) gets a value of 1 and 0 otherwise on the new variable X. >> >> Similarly, I am trying to generate a variable X2 by project_id and org_id that does the following: >> >> if org_id (i) has a new niche value =1, and collaborates with another org_id (j) that has an org_elite value ( <.), then org_id (i) gets a value on the new variable X2 that equals the org_elite value for org_id (j). > > > > * > * 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/ * * 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/

- Prev by Date:
**Re: st: testing the joint significance** - Next by Date:
**Re: st: testing the joint significance** - Previous by thread:
**st: testing the joint significance** - Next by thread:
**st: marginal effect Nested Logit regression** - Index(es):