Statalist


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

st: RE: fitting a gompertz curve, not in the context of survival analysis


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: fitting a gompertz curve, not in the context of survival analysis
Date   Fri, 29 May 2009 11:20:52 +0100

In addition to other comments: 

I'd be interested to know the Laird reference implied but not given
here. 

I don't know of an add-on, but no matter: you shouldn't need one. An
easy alternative should be to try -nl- with the equation specified on
the command line. 

But I'm puzzled by your intent on various levels. 

1. Growth models in biology and elsewhere that I know about presuppose a
natural time origin, i.e. a date of birth or creation. It's not obvious
to me from the rudimentary economics I know that there is an equivalent
for your problem. 

2. Otherwise put, this curve goes X(0) at time 0, but why is one data
point privileged over the others? 

Nick 
n.j.cox@durham.ac.uk 

Dan Waldo

I am trying to test whether government revenues as a percentage of GDP
can be fit over time with a Gompertz curve -- especially to determine
(if the fit is appropriate) the limiting value.

The help I have found comes in the form of survival analysis models,
which I have trouble reconciling to the model I am trying to fit.

Is there a Stata add-on that fits Laird's tumor-growth variant of the
Gompertz? This would be specified as:

X(t)= K exp(log(X(0)/K)exp(-at))

where X is known and K and a are estimated.


*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index