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]
st: saving estimated parameters with -parmby-
From
Schöler, Lisa <[email protected]>
To
"[email protected]" <[email protected]>
Subject
st: saving estimated parameters with -parmby-
Date
Thu, 2 Feb 2012 16:37:35 +0000
Dear Statalist,
I have a dataset with many firms and many observations of a firm. Now I need to run a nonlinear least-squares estimation.
After running the nonlinear least-squares estimation, I would like to save the estimated parameters. So I found suggestions for parmest, in my case parmby because I have different group_ids.
But what I need is that Stata saves each estimate (i.e., b0-b3) in a row, for example
group_id adopters var2 ... t b0 b1 b2 b3
1 500 . . 1 0.3 5 88 1
1 501 2 0.3 5 88 1
1 502 3 0.3 5 88 1
1 503 4 0.3 5 88 1
1 504 5 0.3 5 88 1
1 505 6 0.3 5 88 1
2 100 1 5 4 0.2 0.8
2 106 2 5 4 0.2 0.8
2 108 3 5 4 0.2 0.8
3 250 1 1 0.5 0.8 5
3 297 2 1 0.5 0.8 5
3 300 3 1 0.5 0.8 5
This is the command I use (Stata 11)
. parmby "nl (adopters= {b0}/(1+exp({-b1}-{b2}*t))+{b3}*t^2)",by(group_id) saving(nl_coefficients,replace)
This works but the estimated parameters are saved like this
group_id parmseq estimate
1 b0 238421.79
1 -b1 730.78458
1 b2 21.785424
1 b3 3.0373359
2 b0 8437.5189
2 -b1 41.075473
2 b2 6.0825979
2 b3 -.24629756
Is there an option that I am missing?
Thank you
Lisa
Lisa Schöler
Professur für BWL, insb. Electronic Commerce
Goethe-Universität Frankfurt
Grüneburgplatz 1
60629 Frankfurt a.M.
Phone: ++49 69 798 34632
Fax: ++49 69 798 35001
Email: [email protected]
URL: www.ecommerce.wiwi.uni-frankfurt.de
_________________________________
Get serious about better marketing decisions - apply now!
http://www.goethe-master.de
Lisa Schöler
Professur für BWL, insb. Electronic Commerce
Goethe-Universität Frankfurt
Grüneburgplatz 1
60629 Frankfurt a.M.
Phone: ++49 69 798 34632
Fax: ++49 69 798 35001
Email: [email protected]
URL: www.ecommerce.wiwi.uni-frankfurt.de
_________________________________
Get serious about better marketing decisions - apply now!
http://www.goethe-master.de
*
* 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/