summary.nls
Summary of an nls Model Object
Description
This is an S3 method for the function summary for objects inheriting from class nls.
Usage
summary.nls(object, correlation = FALSE, symbolic.cor = FALSE, ...)
Arguments
| object |
A fitted non-linear least-squares model inheriting from class nls.
|
| correlation |
a logical flag, to specify if the correlation matrix of the estimated
parameters will be returned and printed. This argument is not used at
present.
|
| symbolic.cor |
a logical flag, if TRUE, the function will print the correlations
in a symbolic form rather than as numbers. This argument is not used at
present.
|
Value
returns a list of class summary.nls
whose components call, convInfo, control, and na.action are copied
from its object argument and whose other components are
| formula |
the formula component from object.
|
| residuals |
the weighted residuals.
|
| sigma |
the square root of the estimated variance of the random error.
|
| df |
the degrees of freedom.
|
| cov.unscaled |
a matrix of unscaled covariances of the parameter estimates.
|
| coefficients |
a matrix of four columns: estimated coefficient, standard error, t-statistic, and corresponding p-value.
|
| parameters |
same as coefficients.
|
| correlation |
if correlation is TRUE, and if there is a non-zero number of residual degrees of freedom, this component is added, to indicate the correlation matrix corresponding to the cov.unscaled,
|
| symbolic.cor |
if correlation is TRUE, and if there is a non-zero number
of residual degrees of freedom, this component is added with the value
of the argument symbolic.cor.
|
| message |
if object's component call has an algorithm of "port", this component is set with the value copied from object's corresponding component.
|
| ... |
additional arguments passed to or from other methods.
|
Note
This is a method for the function summary() for objects inheriting from class nls.
See summary or summary.default for the general behavior of this function
and for the interpretation of object.
Examples
library(Sdatasets)
fit1 <- nls(circumference ~ A/(1 + exp(-(age-B)/C)), data = Orange,
start = list(A=150, B=600, C=400))
summary(fit1)