如何执行此移动性计算？

发布于 2025-01-30 14:08:19 字数 396 浏览 3 评论 0原文

我有关于公司之间发明家的流动性的数据。对于每个发明家，旧公司和新公司都有多行，每行都呈现她/他在各自公司中获得的专利（请参阅屏幕截图）。我只对此数据集中的两行感兴趣 - 该行包含了旧公司中发明家生产的最后一项专利，而该行包含新公司中的第一份专利。我如何仅保留每个发明家的这两行并删除其余的？该示例中的发明家是38592732，他从科学联盟搬到了阿迪尔。我只想保留62和63行并删除其余的。 - 在Stata或Excel方面的任何帮助将有所帮助。多谢。

原文

分享到QQ

分享到微博

如果你对这篇内容有疑问，欢迎到本站社区发帖提问参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。

发布评论

需要登录才能够评论，你可以免费注册一个本站的账号。

半岛未凉 2025-02-06 14:08:19

可重现的数据集（例如dataEx）将为您提供一个需要较少适应的解决方案，但是下面的示例应显示如何完成。

* Example generated by -dataex-. For more info, type help dataex
clear
input byte inventorID str1 patentID str9 firm int year
1 "A" "old_firm"  2000
1 "B" "old_firm"  2001
1 "C" "new_firm"  2002
1 "D" "new_firm"  2003
1 "E" "new_firm"  2004
1 "F" "new_firm"  2005
2 "G" "old_firm2" 2000
2 "H" "old_firm2" 2001
2 "I" "new_firm2" 2004
2 "J" "new_firm2" 2005
end

* Sort data on inventor and year
sort inventorID year

* Identify last patent as row where inventor is the same on the next row but firm is not
gen lastPatent  = (inventorID == inventorID[_n+1] & firm != firm[_n+1])
* Identify last patent as row where inventor is the same on the previous row but firm is not
gen firstPatent = (inventorID == inventorID[_n-1] & firm != firm[_n-1])

* Keep only rows categorized as first or last patent
keep if lastPatent == 1 | firstPatent == 1

A reproducible data set (using for example dataex) would have given you a solution that would have required less adaptation, but the example below should show you how it can be done.

* Example generated by -dataex-. For more info, type help dataex
clear
input byte inventorID str1 patentID str9 firm int year
1 "A" "old_firm"  2000
1 "B" "old_firm"  2001
1 "C" "new_firm"  2002
1 "D" "new_firm"  2003
1 "E" "new_firm"  2004
1 "F" "new_firm"  2005
2 "G" "old_firm2" 2000
2 "H" "old_firm2" 2001
2 "I" "new_firm2" 2004
2 "J" "new_firm2" 2005
end

* Sort data on inventor and year
sort inventorID year

* Identify last patent as row where inventor is the same on the next row but firm is not
gen lastPatent  = (inventorID == inventorID[_n+1] & firm != firm[_n+1])
* Identify last patent as row where inventor is the same on the previous row but firm is not
gen firstPatent = (inventorID == inventorID[_n-1] & firm != firm[_n-1])

* Keep only rows categorized as first or last patent
keep if lastPatent == 1 | firstPatent == 1

回复收藏 0 原文

~没有更多了~