Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | ipianah nic <ipianahnic@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: RE: From: ipianah nic <ipianahnic@gmail.com> |
Date | Fri, 11 Feb 2011 12:08:22 +0000 |
Nick, my variables are genome and csf, and suppose I use glm/nl what will be the model that i can use to extrapolate data so that i get predicted values for where csf values are missing as below genome csf > 1 2020 > 2 1747 > 3 1667 > 4 1608 > 5 1578 > 6 1552 > 7 1540 > 8 1526 > 9 1494 > 10 1484 > 11 1476 > 12 1460 > 13 1455 > 14 1449 > 15 1442 > 16 1426 > 17 1418 > 18 1412 > 19 1390 > 20 > 21 > 22 > 23 > 24 Ipianah On 2/11/11, Nick Cox <n.j.cox@durham.ac.uk> wrote: > You decide whether it is more appropriate to use -nl-, -regress- after > transformations, or -glm- with a log link. That's (at least) three to choose > from. There is no answer free of assumptions about the most likely kind of > error term. > > Also, if this is student homework, you got more help than you deserved. Real > data that people care about don't arrive as Y and X. > > Nick > n.j.cox@durham.ac.uk > > ipianah nic <ipianahnic@gmail.com> > > how do I fit an exponential regression curve and extrapolate the data > using sample data below; > genome csf > 1 2020 > 2 1747 > 3 1667 > 4 1608 > 5 1578 > 6 1552 > 7 1540 > 8 1526 > 9 1494 > 10 1484 > 11 1476 > 12 1460 > 13 1455 > 14 1449 > 15 1442 > 16 1426 > 17 1418 > 18 1412 > 19 1390 > 20 1389 > 21 1384 > 22 1380 > 23 1380 > 24 1365 > > > * > * 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/