[R] Problem with LM

Richard M. Heiberger rmh @ending from temple@edu
Wed Dec 19 01:10:10 CET 2018


## This example, with your variable names, works correctly.

z2 <- data.frame(y=1:5, x=c(1,5,2,3,5), x2=c(1,5,2,3,5)^2)
z2
class(z2)
length(z2)
dim(z2)

lm(y ~ x + x2, data=z2)

## note that that variable names y, x, x2 are column names of the
## data.frame z2

## please review the definitions and examples of data.frame in ?data.frame
## also the argument requirements for lm in ?lm

On Tue, Dec 18, 2018 at 6:32 PM rsherry8 <rsherry8 using comcast.net> wrote:
>
> The values read into z2 came from a CSV file. Please consider this R
> session:
>
>  > length(x2)
> [1] 1632
>  > length(x)
> [1] 1632
>  > length(z2)
> [1] 1632
>  > head(z2)
> [1] 28914.0 28960.5 28994.5 29083.0 29083.0 29083.0
>  > tail(z2)
> [1] 32729.65 32751.85 32386.05 32379.75 32379.15 31977.15
>  > lm ( y ~ x2 + x, z2 )
> Error in eval(predvars, data, env) :
>    numeric 'envir' arg not of length one
>  > lm ( y ~ x2 + x, as.data.frme(z2) )
> Error in as.data.frme(z2) : could not find function "as.data.frme"
>  > lm ( y ~ x2 + x, as.data.frame(z2) )
> Error in eval(predvars, data, env) :
>    numeric 'envir' arg not of length one
> lm(formula = y ~ x2 + x, data = as.data.frame(z2))
>
> Coefficients:
> (Intercept)           x2            x
>   -1.475e-09    1.000e+00    6.044e-13
>
>  > min(z2)
> [1] 24420
>  > max(z2)
> [1] 35524.85
>  > class(z2)
> [1] "numeric"
>  >
>
> where x is set to x = seq(1:1632)
> and x2 is set to x^2
>
> I am looking for an interpolating polynomial of the form:
>      Ax^2 + Bx + C
> I do not think the results I got make sense. I believe that I have a
> data type error.  I do not understand why
> I need to convert z2 to a data frame if it is already numeric.
>
> Thanks,
> Bob
>
> ______________________________________________
> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.



More information about the R-help mailing list