phenopix (version 2.4.5)

SplineFit: 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))

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

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. greenProcess sf is automatically calculated.

Author

Gianluca Filippa <[e-mail]>

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.

Examples

if (FALSE) {
data(bartlett2009.filtered)
fitted <- SplineFit(bartlett2009.filtered, uncert=TRUE, nrep=50)
}