require more pixels to estimate the variance in the emission-line amplitude #137
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
… 8 pixels to compute statistics
Pull Request Test Coverage Report for Build 5706953194Warning: This coverage report may be inaccurate.This pull request's base commit is no longer the HEAD commit of its target branch. This means it includes changes from outside the original pull request, including, potentially, unrelated coverage changes.
Details
💛 - Coveralls |
|
Drilling into 
To get the variance in the continuum around the line, I estimate the "robust" sigma of the pixels in the continuum as the 75 percentile minus the 25 percentile divided by 1.349: If I instead use a simple 3-sigma-clipped standard deviation the S/N for [OIII] 4363 in this object reduces to ~9.9, which is pretty consistent with the "robust" sigma: 
But I'm still not quite satisfied in these results. |
…the line-amplitude after resampling and convolution
…itial values, so do not assume that they have not been optimized and so do not drop them (this is drop2)
|
@yuvoonng here is the latest fit for the object we've been discussing and below that are the full results. You can see that, as expected, the new line-amplitudes are more in-line with what the data show. You may also notice that the fluxes have changed, too, due to---it turns out---a different bug, which I'll announce more broadly to the group. 
|
|
@moustakas Nice to see that! |
|
The change reported in #127 (comment) leads to (much) larger flux uncertainties so the flux S/N distribution now is more sensible: 
|
…tegration area will be 2*3*linesigma_ang = 12 pixels
…n ContinuumTools class
…he measured not reduced chi-squared
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
@yuvoonng
This PR addresses #124 by requiring more pixels when estimating the variance in the continuum around narrow emission lines, which more realistically helps identify lower signal-to-noise ratio lines. Previously, I required a minimum of just 3 pixels around each emission line; now I require 8.
Using the sample of 40 objects in #124, here are the new S/N distributions for [OIII] 4363:
And here are some numbers (where
v1==Iron/v1.0andv2==this branch.Note that the final numbers may be affected by whatever we decide to do about #127, but nevertheless I'm planning to merge this PR as "the right thing to do" regardless.