# Re: [R] formula

Date: Fri 16 Jul 2004 - 02:06:07 EST

If you want to fit "y = a + bx", then you use "lm(y ~ x)" instead of "lm(y ~ A + bx)". 'A' is not a parameter but coefficient and you do not need to specify coefficients, which is what the linear model is trying to do anyway ! See the details section of help("formula").

> x <- 1:5
> y <- c(0, 1.0, 1.7, 2.0, 2.1)
> lm(x ~ y)

Call:
lm(formula = x ~ y)

Coefficients:

```(Intercept)            y
0.6828       1.7038

```

If A was already defined, and you are trying to multiply the y-values, then use the I() operator. This protects the term/calculation and inhibits the usual term interpretation in linear models.

Call:
lm(formula = x ~ I(2 * y))

Coefficients:

```(Intercept)     I(2 * y)
0.6828       0.8519

```

PS : I think there is a typo in the y input as there is no such number as 2.1.4

On Thu, 2004-07-15 at 16:28, solares@unsl.edu.ar wrote:
> Hi, i 'dont understand how to take a general formula, view this:
>
> x<-1:5
> y<-c(0,1,1.7,2,2.1.4)
> dummy<-data.frame(x=x,y=y)
> formula<-"y~A*log(x)/log(2)"
> formu<-as.formula(formula)
> fm<-lm(formu,data=dummy)
>
> but A is the parameter of fitting, why is this?Thanks Ruben
>
> ______________________________________________
> R-help@stat.math.ethz.ch mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-help