Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
| From | "Kate Rohrbaugh" <krohrbaugh@ipaglobal.com> |
| To | <statalist@hsphsun2.harvard.edu> |
| Subject | st: !!NOSIG!! CI for nonlinear estimation |
| Date | Tue, 24 Aug 2010 14:02:28 -0400 |
Greetings Statalist:
We have Stata 9.2, and would like to figure out 95% CIs for predicted values around a beta distribution (i.e., cash flow curves over the life of a project), if it is even possible. Apologies that I cannot provide dataset or actual data since our work is confidential.
Here's the Stata command:
nl ( _cexpend = ibeta({alpha=1},{beta=1}, _cphase))
where _cexpend is the cumulative expenditure (in percent of total)
_cphase is the cumulative phase (in percent of total)
We've looked at predictnl and nlcom, but need some guidance on applying the results from these commands, or other guidance on what we should be doing.
Thanks for any ideas!
Kate
Kate Rohrbaugh
Research Team Leader
Research and Special Studies
Independent Project Analysis, Inc.
44426 Atwater Drive
Ashburn, VA 20147
direct - 703.726.5465
front desk - 703.729.8300
fax - 703.729.8301
email - krohrbaugh@ipaglobal.com
website - www.ipaglobal.com
*
* 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/