pppvalue: Pointwise Adjusted Permutation P-values (PPPvalue)
In wtagr/pppvalue: Pointwise adjusted Permutation P-value (PPPvalue)

Description Usage Arguments Details Value Note Author(s) References See Also Examples

View source: R/pppvalue.R

Perform three distribution-free adjustments for pointwise p-values based on envelope/graphical tests. The current version is made available for testing.

1	pppvalue(realvec, permat, method = "all")

`realvec`	A vector of observed data.
`permat`	A matrix of simulated/permuted data, the `nrow` of which must be the same as the length of `realvec` and the `ncol` of which denotes the number of the simulation or permutation.
`method`	A charactor option specifying three adjustments or all of these, for `'single-step'`, `'step-down'`, `'ERL'` and `'all'`. The default is `'all'`. See details.

There are three distribution-free p-value adjustments offered in this package:

'single-step': the single-step adjusted p-values with respect to the GET-package;
'step-down': the step-down adjusted p-values with respect to the step-down procedure (Westfall and Young, 1993; Ge et al., 2003; Romano and Wolf, 2016);
'ERL': the adjusted p-values based on envelopes of the extreme rank length (ERL).

Please see Xu and Reiss (2019) for more details.

Estimated pointwise adjusted p-values.

'ERL' and 'all' also offer the pointwise raw p-values w.r.t ranks and the p-value of GET in the default setting;
The rank adopts the default tie.method;
These adjustments would not consider the ties situations currently.

Meng Xu mxu@campus.haifa.ac.il, Philip Reiss

Xu, M., and Reiss, P. T. (2019). Distribution-Free Pointwise Adjusted P-Values for Functional Hypotheses. https://arxiv.org/abs/1912.00360.

The plot function pppvalue.plot. Other useful resources on Global Evelope Test refer to the GET package (global_envelope_test).

## Not run: 
set.seed(123)
library(pppvalue)
obs <- rnorm(20,mean=2, sd=2)
perms <- matrix(rnorm(20*999),20)
pvals <- pppvalue(obs, perms)
pppvalue.plot(pvals)

## End(Not run)