Harmonic Fit to Time Series — harmonicfit • iAR

This function fit an k-harmonic function to time series data.

Usage

harmonicfit(x, ...)

Arguments

x

An object of class `utilities`.

...

Additional arguments for pairing time series:

data: A data frame with three columns corresponding to the time, values, and standard errors of the irregularly observed time series.
f1: frequency (1 / period) of the time series.
nham: Number of harmonic components in the model.
weighted: logical; if true, a weighted least squares (WLS) estimation is performed using weights based on the standard deviations of the errors. Default is 'FALSE'.
remove_trend: logical; if true, the linear trend of time series will be removed before the the harmonic model is fitted.

Value

An object of class `utilities` with the slots:

fitted_values: Fitted values from the harmonic model.
residuals: Residuals from the harmonic model.
coef: Estimated coefficients of the harmonic model.
summary: A summary object containing detailed model information.

Details

The function fits a harmonic regression model with 'nham' components to the input time series, optionally removing a linear trend and allowing for weighted estimation when standard errors are available.

References

Elorrieta F, Eyheramendy S, Palma W, Ojeda C (2021). “A novel bivariate autoregressive model for predicting and forecasting irregularly observed time series.” Monthly Notices of the Royal Astronomical Society, 505(1), 1105-1116. ISSN 0035-8711, doi:10.1093/mnras/stab1216 , https://academic.oup.com/mnras/article-pdf/505/1/1105/38391762/stab1216.pdf.

Examples

data(clcep)
f1=0.060033386
o1=iAR::utilities()
o1<-phase(o1,data=clcep,f1=f1)
#results$R2
#results$MSE
#results=harmonicfit(file=clcep[,1:2],f1=f1,nham=3)
#results$R2
#results$MSE
#results=harmonicfit(file=clcep[,1:2],f1=f1,weights=clcep[,3])
#results$R2
#results$MSE