Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Chris Min <cmsk0109@yahoo.com> |
To | "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu> |
Subject | Re: st: How to set a range from 0 to positive infinity in calculating integrals? |
Date | Mon, 10 Oct 2011 12:48:10 -0700 (PDT) |
Dear Maarten, Thank you very much for your help -- I find your explanation very useful. But, I did mean "normal()" in my example (i.e., I need to calculate integrals of cumulative normal), not normalden(). In that case, would your explanation still apply to my example? Thanks, Chris ----- Original Message ----- From: Maarten Buis <maartenlbuis@gmail.com> To: statalist@hsphsun2.harvard.edu Cc: Sent: Monday, October 10, 2011 1:58 AM Subject: Re: st: How to set a range from 0 to positive infinity in calculating integrals? On Sun, Oct 9, 2011 at 10:06 PM, Chris Min wrote: > I want to use the command "integ" to calculate for the > following integral of y over x ranging from 0 to positive > infinity. But I don't know how to set a range for x (i.e., > from 0 to positive infinity). If anyone can help me with > the second line of my coding, that would be great. > > set obs 100 > range x 0 ? > gen y=1-normal(ax-b) /* where a and b are scalars */ > integ y x, g(integral) In a strict sense you cannot use -integ- to go to positive infinity. It takes the x and y values in your dataset to be your function and calculates the area underneath the curve for that function. Since you cannot have positve infinity in a dataset, you cannot do what you want. However, typically, a suitably large number can give you a very good approximations, and these numbers don't have to be huge: in case of a standard normal/Gaussian distribution you could use an upper bound of 4 as a reasonable approximation for positive infinity... In your example I assume you want to use -normalden()- instead of -normal()-, as the latter is already the cumulative density function. Hope this helps, Maarten -------------------------- Maarten L. Buis Institut fuer Soziologie Universitaet Tuebingen Wilhelmstrasse 36 72074 Tuebingen Germany http://www.maartenbuis.nl -------------------------- * * 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/