SplineFit {phenopix} | R Documentation |
A function to fit a smoothed spline to Phenological Data
Description
A function to fit a smoothed spline to Phenological Data
Usage
SplineFit(ts, uncert = FALSE, nrep = 100, df.factor = 0.05,
ncores='all', sf=quantile(ts, probs=c(0.05, 0.95), na.rm=TRUE))
Arguments
ts |
A ts or zoo object with gcc data. index(ts) must be numeric days of year (doys) |
uncert |
Should uncertainty be estimated? |
nrep |
Number of relications to estimate uncertainty, defaults to 100. |
df.factor |
Defaults to 0.05, it is multiplied by length(ts) to generate degrees of freedom for the spline fitting. The higher the number of data, the higher should be df factor. For a complete year of data (i.e. length(ts)=365) the default value is optimum. |
ncores |
Unused argument for compatibility |
sf |
Scaling factors required to normalize the data prior to the fitting. If the function is called by e.g. |
Details
This function fits a smoothed spline to the data. Df for smoothing are set at 0.05*length(ts) by default and df.factor can be modified. Uncertainty is estimated by changing the degrees of freedom of the spline. In particular a sequence from 0.01 and df.factor, of length nrep is used as varying degrees of freedom for the spline fitting.
Value
A list containing the following items.
fit |
A list with fitted values and an object named 'params' set to NULL, for simmetry with other fittings |
uncertainty |
A list containing a zoo data.frame with the uncertainty predicted values, and an object named 'params' set to NULL, for simmetry with other fittings |
Author(s)
Gianluca Filippa <gian.filippa@gmail.com>
Examples
## Not run:
data(bartlett2009.filtered)
fitted <- SplineFit(bartlett2009.filtered, uncert=TRUE, nrep=50)
## End(Not run)