Convert WaveX to PLRedNoise and DMWaveX to PLDMNoise
#1694
Merged
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
abhisrkckl
changed the title
WaveX/DMWaveX componentsWaveX to PLRedNoise and DMWaveX to PLDMNoise
Codecov ReportAttention:
Additional details and impacted files@@ Coverage Diff @@
## master #1694 +/- ##
==========================================
+ Coverage 68.78% 68.96% +0.18%
==========================================
Files 105 105
Lines 24643 24696 +53
Branches 4409 4409
==========================================
+ Hits 16950 17031 +81
+ Misses 6588 6560 -28
Partials 1105 1105 ☔ View full report in Codecov by Sentry. |
|
Red noise conversion example (simulation): |
|
DM noise conversion example (simulation): |
abhisrkckl
changed the title
WaveX to PLRedNoise and DMWaveX to PLDMNoiseWaveX to PLRedNoise and DMWaveX to PLDMNoise
kerrm
reviewed
Jan 11, 2024
kerrm
reviewed
Jan 11, 2024
abhisrkckl
added
the
awaiting review
|
This looks good, but it looks like the fit comparisons are only done in the example and not in the tests. Are additional tests possible? |
|
@dlakaplan I have added quantitative tests. |
dlakaplan
reviewed
Jan 19, 2024
dlakaplan
reviewed
Jan 19, 2024
src/pint/utils.py
Outdated
|
|
||
| mlnlike = _get_wx2pl_lnlike(model, "WaveX", ignore_fyr=ignore_fyr) | ||
|
|
||
| gamma_val, log10_A_val = minimize(mlnlike, [4, -13], method="Nelder-Mead").x |
There was a problem hiding this comment.
are there any checks for whether or not this has actually converged?
There was a problem hiding this comment.
added convergence checks
dlakaplan
reviewed
Jan 19, 2024
|
@dlakaplan Can you please merge this if it looks fine? |
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.
WaveXandPLRedNoiseare both Fourier representations of achromatic red noise. The difference is thatWaveXis a deterministic model with free Fourier coefficients whereasPLRedNoiseis a GP model where the Fourier coefficients have been analytically marginalized. TheWaveXcoefficients can be converted toPLRedNoisespectral parameters (amplitude and spectral index) by maximizing the following likelihood function:Here, a_j and b_j are the
WaveXcoefficients, epsilon represents their measurement uncertainties, and P(f) is a power law spectrum.This likelihood function has a different statistical interpretation than the analytically marginalized likelihood used for Bayesian analysis (it acts on timing residuals directly). Hence, its results are not expected to be fully consistent with the Bayesian results. Nevertheless, this conversion gives results fairly close to the Bayesian estimates (mostly within the 3-sigma level in the tests I have done).
The inverse
PLRedNoise-->WaveXconversion without using the TOAs is not possible sincePLRedNoisehas less information thanWaveX.A similar relationship and conversion also apply to
DMWaveXandPLDMNoise.Please note that this does not work properly for wideband data due to bug #1660.
Also includes some sourcery refactors.