getHarmMetrics: fits robust linear harmonic model to ts and retunrs the...

In johanez/probaV: Utils for time-series analysis of Proba-V data

Description

Processes Proba-V data for subsequentent use in time-series analysis. Performs Proba-V cleaning and operations with parallel support.

Usage

getHarmMetrics(x, dates = NULL, QC_good = NULL, sig = 0.95,
  n_years = NA, lin_trend = T, order = c(1, 2, 3), robust = F,
  return_model = F, probs = c(0.01, 0.99), ...)

Arguments

x	Numeric or ts Time series
dates	Dates oberservation times if no time series is supplied
QC_good	Integer or Logical. If supplied, values of x that are not 1 or TRUE in QC_good are omitted. Must have same length as x and dd.
sig	Numeric If not NA, coeficients with p-value > sig will be set to 0.
n_years	Integer. If not NULL, additional coeff. for n intra-yearly variation.
lin_trend	Logical. Should a linear trend be modelled?
order	Numeric. First or second order harmonics
robust	Logical Use robust regression from robustbase:lmrob?
probs	Numeric. Vector of two probabilities to compute quantiles.
...	additional args. non implemented.

Value

Numeric vector of the models coefficients and percentiles (length depending on arguments)

See Also

removeDips, and robustbase::lmrob

johanez/probaV documentation built on May 19, 2019, 3:01 p.m.