baytsDD: baytsDD

In jreiche/bayts: Probabalistic approach to combine multiple time series and Bayesian probability updating for near real-time forest change detection

Description

(i) deseasonalising time series using harmonic model fitting , (ii) deriving F and NF pdfs in data driven way and (iii) run bayts to detect forest chang

Usage

baytsDD(tsL = list(NULL, ...), pdfsdL = list(), distNFL = list(),
  start_history = NULL, end_history = NULL, start, end = NULL,
  formulaL = list(), orderL = list(), bwf = c(0.1, 0.9), chi = 0.9,
  PNFmin = 0.5, residuals = FALSE)

Arguments

tsL	list of object(s) of class ts.
pdfsdL	list of msdl object(s) describing the modulation of the sd of F and NF sd(F),sd(NF),mean(NF) (e.g. 2,2,-4)
distNFL	list of "distNF" object(s) describing the mean and sd of the NF distribution in case no data driven way to derive the NF distribution is wanted; default=NULL
start_history	Start date of history period used to model the seasonality and derive F and NF PDFs. Default=NULL (start of input time series)
end_history	End date of history period used to model the seasonality and derive F and NF PDFs. Default=NULL (Start of the monitoring period is used)
start	Start date of monitoring period. Default=NULL (start of input time series)
end	End date of monitoring period. Default=NULL (start of input time series)
formulaL	list of formula for the regression model. The default is "response ~ harmon", i.e., a harmonic season component without trend component: "response ~ trend + harmon"; With trend: Other specifications are possible using all terms set up by bfastpp, i.e., season (seasonal pattern with dummy variables), lag (autoregressive terms), slag (seasonal autoregressive terms), or xreg (further covariates). See bfastpp for details.
bwf	block weighting function to truncate the NF probability; Default=c(0.1,0.9); (c(0,1) = no truncation)
chi	Threshold of Pchange at which the change is confirmed; Default=0.5
PNFmin	threshold of pNF above which the first observation is flagged; Default=0.5
residuals	TRUE = output are time series of deseasonalised "residuals"; FALSE = time series of deseasonalised values
order	list ofnumeric. Order of the harmonic term, defaulting to 3.

Value

List of 7 output paramter. (1) data_tspp: "data_tspp" time series data frame (2) bayts: "bayts" time series data frame (3) pdf: pdfs for forest and non-forest (4) flag: time at which unconfirmed change got flagged; (5) change.flagged: time at which confirmed change got flagged; (6) change.confirmed: time at which change is confirmed; (7) oldflag: time of earlier flagged but not confirmed changes; (8) vchange: vector of time steps from time at which change got flagged until confirmation; (9) vflag: vector of time steps at which unconfirmed change is flaged

Author(s)

Johannes Reiche (Wageningen University)

jreiche/bayts documentation built on Feb. 3, 2021, 1:12 a.m.