Skip to content
GitLab
Projects Groups Topics Snippets
  • /
  • Help
    • Help
    • Support
    • Community forum
    • Submit feedback
  • Sign in
  • s2dv s2dv
  • Project information
    • Project information
    • Activity
    • Labels
    • Members
  • Repository
    • Repository
    • Files
    • Commits
    • Branches
    • Tags
    • Contributor statistics
    • Graph
    • Compare revisions
  • Issues 16
    • Issues 16
    • List
    • Boards
    • Service Desk
    • Milestones
  • Merge requests 2
    • Merge requests 2
  • CI/CD
    • CI/CD
    • Pipelines
    • Jobs
    • Schedules
  • Deployments
    • Deployments
    • Environments
    • Releases
  • Packages and registries
    • Packages and registries
    • Package Registry
    • Terraform modules
  • Monitor
    • Monitor
    • Incidents
  • Analytics
    • Analytics
    • Value stream
    • CI/CD
    • Repository
  • Wiki
    • Wiki
  • Snippets
    • Snippets
  • Activity
  • Graph
  • Create a new issue
  • Jobs
  • Commits
  • Issue Boards
Collapse sidebar
  • Earth SciencesEarth Sciences
  • s2dvs2dv
  • Issues
  • #79
Closed
Open
Issue created Oct 10, 2022 by aho@ahoMaintainer

Unify the argument names and output regarding significance test

As discussed in the previous user meeting, we plan to make the inputs and outputs regarding significance test in all s2dv functions (if applicable) consistent. The changes include:

  1. Change argument “conf.lev” to “alpha”
    E.g., conf.lev = 0.95 in Corr() will be deprecated and be replaced by new argument alpha = 0.05.

  2. Return “sign” or “p.val”, (and “conf.low”, "conf.upper" if they exist already)
    If alpha = NULL, return "p.val"; if alpha has value, return "sign" (logical value, means significance (T) or not (F))

Functions that need these changes: Corr, ACC, RMS, Regression, Spread, Trend
Functions that have these features already: DiffCorr, ResidualCorr

Questions remain:

  1. How to calculate sign (significance or not)? In DiffCorr(), simply compare alpha with p-value; but in ResidualCorr(), the calculation is more complicated.
  2. RPSS() and CRPSS() use RandomWalkTest() for significance test, so they only have "sign" but not "p.val". Is it possible/needed to add p-value?
Edited Oct 11, 2022 by aho
Assignee
Assign to
Time tracking