parameter_estimation.rst 3.58 KB
Newer Older
1 2
Sky Localization and Parameter Estimation
=========================================
3 4

Immediately after one of the :doc:`search pipelines <searches>` reports an
5 6 7
event, sky localization and parameter estimation analyses begin. These analyses
all use Bayesian inference to calculate the posterior probability distribution
over the parameters (sky location, distance, and/or intrinsic properties of the
8 9 10
source) given the observed gravitational-wave signal.

There are different parameter estimation methods for modeled (CBC) and
Leo Pound Singer's avatar
Leo Pound Singer committed
11 12 13 14
unmodeled (:term:`burst`) events. However, in both cases there is a rapid
analysis that estimates only the sky localization, and is ready in seconds, and
a refined analysis that explores a larger parameter space and completes up to
hours or a day later.
15 16 17 18

Modeled Events
--------------

19 20 21
**BAYESTAR** [#BAYESTAR]_ is the rapid CBC sky localization algorithm. It reads
in the matched-filter time series from the :doc:`search pipeline <searches>`
and calculates the posterior probability distribution over the sky location and
22 23 24
distance of the source by coherently modeling the response of the
gravitational-wave detector network. It explores the parameter space using
Gaussian quadrature, lookup tables, and sampling on an adaptively refined
25 26
:term:`HEALPix` grid. The sky localization takes tens of seconds and is
included in the preliminary alert.
27 28 29 30 31

**LALInference** [#LALInference]_ is the full CBC parameter estimation
algorithm. It explores a greatly expanded parameter space including sky
location, distance, masses, and spins, and performs full forward modeling of
the gravitational-wave signal and the strain calibration of the
Leo Pound Singer's avatar
Leo Pound Singer committed
32 33
gravitational-wave detectors. It explores the parameter space using
:term:`MCMC` and nested sampling. For all events, there is an automated
34 35 36
LALInference analysis that uses the least expensive CBC waveform models and
completes within hours and may be included in a subsequent alert. More
time-consuming analyses with more sophisticated waveform models are started at
Leo Pound Singer's avatar
Leo Pound Singer committed
37
the discretion of human analysts, and will complete days or weeks later.
38 39 40 41 42

Unmodeled Events
----------------

**cWB**, the burst :doc:`search pipeline <searches>`, also performs a rapid
43
sky localization based on its coherent reconstruction of the gravitational-wave
44
signal using a wavelet basis and the response of the gravitational-wave
45 46
detector network [#cWBLocalization]_. The cWB sky localization is included in
the preliminary alert.
47

48
Refined sky localizations for unmodeled bursts are provided by two algorithms
Leo Pound Singer's avatar
Leo Pound Singer committed
49
that use the same :term:`MCMC` and nested sampling methodology as LALInference.
50 51 52 53 54
**LALInference Burst (LIB)** [#oLIB]_ models the signal as a single
sinusoidally modulated Gaussian. **BayesWave** [#BayesWave]_ models the signal
as a superposition of wavelets and jointly models the background with both a
stationary noise component and glitches composed of wavelets that are present
in individual detectors.
Leo Pound Singer's avatar
Leo Pound Singer committed
55

Leo Pound Singer's avatar
Leo Pound Singer committed
56
.. |cqg| replace:: *Class. Quantum Grav.*
Leo Pound Singer's avatar
Leo Pound Singer committed
57 58 59 60
.. |prd| replace:: *Phys. Rev. D*

.. [#BAYESTAR]
   Singer, L. P., & Price, L. R. 2016, |prd|, 93, 024013.
61
   :doi:`10.1103/PhysRevD.93.024013`
Leo Pound Singer's avatar
Leo Pound Singer committed
62 63 64

.. [#LALInference]
   Veitch, J., Raymond, V., Farr, B., et al. 2015, |prd|, 91, 042003.
65
   :doi:`10.1103/PhysRevD.91.042003`
Leo Pound Singer's avatar
Leo Pound Singer committed
66 67 68

.. [#cWBLocalization]
   Klimenko, S., Vedovato, G., Drago, M., et al. 2011, |prd|, 83, 102001.
69
   :doi:`10.1103/PhysRevD.83.102001`
Leo Pound Singer's avatar
Leo Pound Singer committed
70 71 72

.. [#oLIB]
   Lynch, R., Vitale, S., Essick, R., Katsavounidis, E., & Robinet, F. 2017, |prd|, 95, 104046.
73
   :doi:`10.1103/PhysRevD.95.104046`
Leo Pound Singer's avatar
Leo Pound Singer committed
74 75 76

.. [#BayesWave]
   Cornish, N. J., & Littenberg, T. B. 2015, |cqg|, 32, 135012.
77
   :doi:`10.1088/0264-9381/32/13/135012`