<>
Here we go:
***
clear*
input countyid/*
*/ year entry exit
1 1983 35 33
1 1984 37 36
1 1985 38 39
2 1983 120 125
2 1984 122 127
2 1985 125 130
end
compress
list, noobs/*
*/ sepby(countyid)
bys countyid (year): /*
*/egen baseentry=total((year==1984)*entry)
bys countyid (year): /*
*/egen baseexit=total((year==1984)*exit)
bys countyid (year):/*
*/gen entry_index=100*entry/ baseentry
bys countyid (year):/*
*/gen exit_index=100*exit/ baseexit
list, noobs/*
*/ sepby(countyid)
***
HTH
Martin
_______________________
----- Original Message -----
From: "Manfred Fleischer" <[email protected]>
To: <[email protected]>
Sent: Monday, June 15, 2009 8:39 PM
Subject: Re: st: Re: Generating index numbers with panel data in long format
Hi Martin,
Thank you for the hint. I will work through the suggested SJ article.
Sorry, that I have not included the index numbers in the example. Here
they are using 1984 as the base year ...
county-id year entry exit entry_index exit_index
1 1983 35 33 94.6 91.7
1 1984 37 36 100.0 100.0
1 1985 38 39 102.7 108.3
etc.
2 1983 120 125 98.4 98.4
2 1984 122 127 100.0 100.0
2 1985 125 130 102.5 102.4
etc.
Thanks,
Manfred
2009/6/15 Martin Weiss <[email protected]>
<>
Could you give us a hint what the calculation of the "index number" would
look like in your example? I have a feeling that
http://www.stata-journal.com/sjpdf.html?articlenum=pr0004
will be the solution, but just write down what your result should look
like...
HTH
Martin
_______________________
----- Original Message ----- From: "Manfred Fleischer"
<[email protected]>
To: <[email protected]>
Sent: Monday, June 15, 2009 7:57 PM
Subject: st: Generating index numbers with panel data in long format
Hi,
Would anyone know whether there exists a possibility to generate index
numbers over time with panel data stored in long format.
The dataset consists of new firm entry and exit rates for German
counties from 1983 to 2004. It looks like this
county-id year entry exit
1 1983 35 33
1 1984 37 36
1 1985 38 39
etc.
2 1983 120 125
2 1984 122 127
2 1985 125 130
etc.
Now I would like to generate index numbers by county for entry and
exit e.g. with 1984 as the base year. I have looked for egen and
time-series commands, but I couldn't find a solution. Maybe the only
solution is to reshape the file into the wide format, compute the
index numbers there and then return with the generated index numbers
in the long format. But that's what I don't know.
Any help is very much appreciated.
Thanks,
Manfred
----------------------------------------------------------------
Manfred Fleischer
D-52159 Roetgen-Rott (Kreis Aachen), Germany
Page: http://www.wzb.eu/mp/mps/people/manfred_fleischer.en.htm
*
* 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/
