Uploaded image for project: 'Data Management'
  1. Data Management
  2. DM-24762

Add option to PTC task to correct for sigma clipping bias

    Details

    • Type: Story
    • Status: To Do
    • Resolution: Unresolved
    • Fix Version/s: None
    • Component/s: cp_pipe
    • Labels:
      None

      Description

      When sigma-clipping to calculate the mean and variances, biases are potentially introduced at both the high and low flux ends, and should be corrected for somehow.

      It's not clear how to do this well, but this ticket notes at least considering how we should approach it.

        Attachments

          Activity

          Hide
          nmondrik Nicholas Mondrik added a comment -

          To be pedantic, I will note that bias is introduced in the variance estimate at all flux levels (not just high or low), but bias is introduced in the mean estimate only when the distribution is significantly non-symmetric (high- or low-flux, as Merlin said).

          Show
          nmondrik Nicholas Mondrik added a comment - To be pedantic, I will note that bias is introduced in the variance estimate at all flux levels (not just high or low), but bias is introduced in the mean estimate only when the distribution is significantly non-symmetric (high- or low-flux, as Merlin said).
          Hide
          mfisherlevine Merlin Fisher-Levine added a comment -

          That is indeed exactly what I meant to say

          Show
          mfisherlevine Merlin Fisher-Levine added a comment - That is indeed exactly what I meant to say

            People

            • Assignee:
              Unassigned
              Reporter:
              mfisherlevine Merlin Fisher-Levine
              Watchers:
              Andrés Alejandro Plazas Malagón, John Swinbank, Merlin Fisher-Levine, Nicholas Mondrik, Seth Digel
            • Votes:
              0 Vote for this issue
              Watchers:
              5 Start watching this issue

              Dates

              • Created:
                Updated:

                Summary Panel