Quantiles from the gaussian version of the test statistics which are used to approximate the critical values for the multiscale test.
compute_quantiles(
t_len,
n_ts = 1,
grid = NULL,
ijset = NULL,
sigma = 1,
deriv_order = 0,
sim_runs = 1000,
probs = seq(0.5, 0.995, by = 0.005),
correction = TRUE,
epidem = FALSE
)Matrix with 2 rows where the first row contains the vector of probabilities (probs) and the second contains corresponding quantiles of the gaussian statistics distribution.
Sample size.
Number of time series analyzed. Default is 1.
Grid of location-bandwidth points as produced by
the function construct_grid or
construct_weekly_grid, list with
the elements 'gset', 'bws', 'gtype'. If not provided,
then the defalt grid is produced and used.
For the construction of the default grid,
see construct_grid.
A matrix of integers. In case of multiple time series, we need to know which pairwise comparisons to perform. This matrix consists of all pairs of indices \((i, j)\) that we want to compare. If not provided, then all possible pairwise comparison are performed.
Value of \(\sqrt{\sigma^2}\). In case of n_ts = 1, \(\sigma^2\) denotes the long-run error variance, and in case of n_ts > 1, \(\sigma^2\) denotes the overdispersion parameter. If not given, then the default is 1.
In case of a single time series analysed, this parameter denotes the order of the derivative of the trend function that is being estimated. Default is 0.
Number of simulation runs to produce quantiles. Default is 1000.
A numeric vector of probability levels \((1-\alpha)\) for which the quantiles are computed. Default is \((0.5, 0.505, 0.51, \ldots, 0.995)\).
Logical variable, TRUE (by default) if we are using \(a_k\) and \(b_k\).
Logical variable, TRUE if we are using dealing with epidemic time trends. Default is FALSE.