tslars(x, ...) ## S3 method for class 'formula':
tslars(formula, data, ...)
## S3 method for class 'default':
tslars(x, y, h = 1, pMax = 3,
sMax = NA, fit = TRUE, s = c(0, sMax), crit = "BIC",
ncores = 1, cl = NULL, model = TRUE, ...)
rtslars(x, ...)
## S3 method for class 'formula':
rtslars(formula, data, ...)
## S3 method for class 'default':
rtslars(x, y, h = 1, pMax = 3,
sMax = NA, centerFun = median, scaleFun = mad,
regFun = lmrob, regArgs = list(),
combine = c("min", "euclidean", "mahalanobis"),
winsorize = FALSE, const = 2, prob = 0.95, fit = TRUE,
s = c(0, sMax), crit = "BIC", ncores = 1, cl = NULL,
seed = NULL, model = TRUE, ...)
as.data.frame) containing the variables in
the model. If not found in data, the variables are taken
from NA (the
default), predictor groups are sequenced as long as there
are twice as many observations as predictor variables.median).mad).lmrob).regFun."min" for taking the minimum weight for each
observation, "euclidean" for wTRUE, the default) or to
simply return the sequence (FALSE)."BIC" for the Bayes information
criterion is implemented.NA, all available processor cores are used. For
each lag length, parallel computimakeCluster.
This is preferred over ncores for tasks that are
parallelized on the Rlevel, in which case .Random.seed), which is
useful because many robust regression functions
(including lmrobfit is FALSE, an integer matrix in which
each column contains the indices of the sequenced
predictor series for the corresponding lag length. Otherwise an object of class "tslars" with the
following components:
tslarsP).model is TRUE).model is
TRUE).coef,
fitted,
plot,
predict,
residuals,
tslarsP, lmrob