Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Robert Drago <robert.drago44@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: cumulative density by age from density var |
Date | Mon, 7 May 2012 12:44:49 -0400 |
Dear Statalist, Apologies if this has already been answered (feel free to point to an earlier thread!). I have American Community Survey data on adults age 16-85, and have added CDC survival probabilities for each individual year of age (i.e. the density, which I label psurv). Starting with some arbitrary year of age, say 25, I am trying to attach to all future years of age the cumulative density from that age forward. Statistically, the problem is simple: assign value psurv for age 25 to age 25, multiply that value by psurv for age 26 and assign to age 26, multiply the result by psurv for age 27 and assign to age 27, and so forth. Here's what I've tried: reduce the data to one observation for each age with "by age, sort; gen nvals = _n==1" I then did a recode and created a new var with psurv=1 for all ages<25, then ran cumul, but cannot get that command to use the values instead of the number of observations per age. I also tried forvalues, as in: gen psurv25 = psurv recode psurv25 (0/1=1) if age<25 forvalues whatage=25/85{ replace psurv25 = psurv25*psurv if age = `whatage' } but if isn't allowed... Any help greatly appreciated, Bob -- Robert Drago, Ph.D. 1803 BIltmore St NW, #702 Washington DC 20009 ph 814-883-9907 robert.drago44@gmail.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/