Understanding waveform modelling

[herwins]

I am currently an undergraduate student working on a project on generating gravitational waveforms and I have a few questions:

1. Is there a preference when modeling eccentric orbits to use a mass ratio close to one? is it done to make computations faster? precisely two eccentric waveform models/calculations ( one, two)where I have noticed this, I understand two instances are not enough to make a conclusion. (I understand
   LIGO detectors are not sensitive enough to detect eccentricity)

2. Many of the parameters estimated for GW events I have seen have relatively high absolute errors as can be seen here for mass and can take a range of values,
   

   

and these estimated parameters keep changing with new revisions (different versions over time) Firstly what causes these revisions? (does it have to do with Bayesian statistics and new data is somehow used?) secondly, given there is a wide range of possible value for the individual parameters does this mean there is set of waveforms with different set of parameters that can define the event?

3. I have been able to generate a waveform using a program described here (uses the generic Implicit Rotating Source model, which does have some simplifications), and have been trying to compare it with waveform models such as SEOBNRv4HM and IMRPhenomD using the PyCBC library and I formed the following plot:
   
   
   
   I have been trying to use the match feature of the PyCBC library. but am wondering how do i convert by numpy array to be used as a time series which i can use in the matching feature?? also i do understand the matching is performed by evaluating psd but its output is just a number what exactly does the number represent??

I’ve noticed these are many questions for a single post! these were just passing through my head and thought this would be a good place to ask some of them.

[jonah]

@herwins Thank you for the great questions!

1. I’m not really an expert on waveform modeling. You could e-mail the authors of a particular study to learn more about it. But, in general, numerical relativity simulations are easier to run with binaries that have nearly equal masses.

2. Some parameters can be more precisely measured, and some parameters are more difficult to measure. Combinations of the mass can typically be better measured than the individual masses. For lower mass systems, like neutron stars, the chirp mass can be measured very well. For larger black holes (around 50 solar masses), the total mass can be measured well.

3. The match between 2 waveforms is defined in equation (29) of the data analysis guide paper.

I hope that helps. Good luck!

1908.11170.pdf

[herwins]

@jonah The data analysis paper is quite informative and I am learning new things daily, some parameters being able to be measured more precisely seems interesting. I appreciate your help !!!