Find the Curve Fit Equation for the smooth.spline function

Find the Curve Fit Equation for the smooth.spline function

In order to evaluate a trend in data the user might take advantage of the smooth.spline function.  How can they get the curve fit equation for the output?

  Unfortunately there is not a way to get the smoothing equation out of the smooth.spline() function.  It is possible to extract a list containing the details of the fit.  This list contains the knot locations, coefficients, etc.  Please see the following example:

---------------------------
> attach(air)
test <- smooth.spline(ozone, temperature)

#Extract the 'fit':
> test$fit
$knot:
[1] 0.0 0.0 0.0 0.0 0.1 0.2 0.2 0.2 0.3 0.3 0.3 0.3 0.3 0.4 0.4 0.4 0.4 0.4 0.4 0.5 0.5
[22] 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.6 0.6 0.6 0.6 0.6 0.6 0.6 0.6 0.7 0.7 0.7 0.7 0.7
[43] 0.7 0.7 0.7 0.7 0.7 0.8 0.8 0.8 0.8 0.8 0.9 0.9 0.9 1.0 1.0 1.0 1.0

$nk:
[1]  55

$min:
[1] 1

$range:
[1] 5

$coef:
[1]  59 156  57  48  80  63  83  60  83  58  85  65  61  82  74  74  67  81  82  64  92
[22]  50  85  76  70  97  53  88  76  86  82  85  91  73  84  87  75  93  87 100  89  90
[43]  84  89  96  86  96  94  86  98  73  99  55 127  81

attr(, "class"):
[1] "smooth.spline.fit"
-------------------------------------------

For additional information on the attributes of the smooth.spline object please type at the command prompt:

> attributes(test)
$names:
[1] "x"        "y"        "w"        "yin"      "lev"      "cv.crit"  "pen.crit"
[8] "df"       "spar"     "fit"      "call"    

$class:
[1] "smooth.spline"


You can extract any of these attributes by using the '$':

> test$spar
[1] 7e-009