How to run a multiple nonlinear regression in Excel or R? I am very inexperienced with R and have only a limited background with Excel but have some data that I need to run a multiple non-linear regression with. What is the best way to do this? With R or Excel? What commands would I use?
 A: In R, you can use the command nls() (see documentation). For example, for a multiple regression with dependent variable $y$, an intercept $a$, and predictors $x1$ and $x2$ with coefficients $b$ and $c$, respectively, and data stored in variable df:
nls(y ~ a + b*x1 + c*x2, data=df, start=list(a=-100, b=.15, c=-.02))
A: in Excel use Solver Add-in. here's how.


*

*add one row with all parameters of the model

*populate your observations in rows, perhaps, one column for dependent variables, and one column per each independent variable.

*add one more column for predicted Y, insert Excel formula to compute the prediction using all independent variables on this row AND all parameters on fixed parameter row

*add one more column with square of the difference between the predicted Y and and dependent Y on the same row

*at the bottom of this column, sum up all squared errors

*start Excel Solver Add-in

*point its target to square errors

*set minimize option

*select the parameter cells to chage

*Run the thing, it'll change parameters until the SSE is minimized


there's a ton of references in internet, such as this one
