5

I need some assistance with a nonlinear adjust.

I am trying to make a mathematical model that describes the rate of silicic acid escaping from an underwater sediment. For theoretical reasons, the equation resulting from the nonlinear regression must be exponential and have the form $${Y} = A + Be^{CX},$$ in which $A$, $B$ and $C$ are constants, $X$ is an independent predictor (time), and $Y$ is a dependent variable (concentration).

I lack but the basic understanding of statistical software, so I have just tried to make it on Excel. The problem is that there I can only adjust my data to equations like $Y = Ae^{Bx}$, that are without a constant.

I know for sure that R with a specific package will work just fine for this problem, but I cannot find any tutorials to orientate me. I guess that there is a really easy answer for this almost trivial question, but for now it is eluding me. Thank you very much for your time and help!

MarianD
  • 1,493
  • 2
  • 8
  • 17
user73298
  • 51
  • 1
  • [This tutorial here](http://socserv.socsci.mcmaster.ca/jfox/Books/Companion/appendix/Appendix-Nonlinear-Regression.pdf) could help. – COOLSerdash Apr 11 '15 at 21:03
  • Worked examples appear (*inter alia*) at http://stats.stackexchange.com/questions/13053, http://stats.stackexchange.com/questions/63978, http://stats.stackexchange.com/a/71097/919 – whuber Apr 13 '15 at 16:36
  • 1
    You can fit nonlinear least squares also in Excel using the optimization facilities, but they won't give you standard errors of coefficients and so forth. – Glen_b May 15 '15 at 03:53

0 Answers0