Hi,
I have the unbalanced panel data like this:
Id yr var1 var2 var3
1 1991 2 3 4
1 1992 32 32 22
1 1993 32 2 23
2 1992 21 32 32
2 1993 32 49 88
3 1991 22 32 34
3 1992 32 87 76
3 1993 33 32 43
4 1991 3 44 33
5 1993 32 33 32
. .. .. .. ..
I would like to transform the data into new one (sorted by the number of
years that we observed the ID i)
Id yr var1 var2 var3
4 1991 3 44 33
5 1993 32 33 32
2 1992 21 32 32
2 1993 32 49 88
1 1991 2 3 4
1 1992 32 32 22
1 1993 32 2 23
3 1991 22 32 34
3 1992 32 87 76
3 1993 33 32 43
I know how to transform to this data in Stata:
sort ID
by ID: egen pd= count(var1)
sort pd
by pd: egen Np= count(pd)
replace Np= Np/pd
gsort pd ID year
But I don't know how to write a program can do this transformation and give
me pd and Np into vector pd[.,1], Np[.,1],
(in this example, pd[]=(1 2 3)' and Np[]=(2 1 2)').And how I can use
pd[.,1], Np[.,1] to partition
the whole transformed matrix [Id yr var1 var2 var3] into blocks(i) [Id yr
var1 var2 var3] with rows = pd[i,1]*Np[i,1]?
Can someone help me on this problems?
Thank you so much,
Minh
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
