Explore advanced gravitational-wave analyses beyond the core workshop curriculum.
This section collects optional notebooks that extend the workshop into advanced areas of gravitational-wave astronomy.
These notebooks cover production-level parameter estimation with LALInference, population inference using published catalogs, and continuous-wave searches based on the Frequency-Hough transform.
The objective is to provide exposure to specialized techniques used in current gravitational-wave research and to highlight directions for deeper exploration.
Reproduces a production-style parameter estimation workflow using the LALInference software employed during the first three observing runs.
Introduces hierarchical Bayesian inference for studying the population properties of compact-object mergers.
Demonstrates the Frequency-Hough method for detecting long-lived, nearly monochromatic gravitational-wave signals.
By working through these notebooks, you will learn how to:
- Configure and launch
LALInferenceanalyses. - Analyze posterior samples from multiple events.
- Infer population-level parameters such as mass distribution slopes.
- Construct peakmaps from time-frequency data.
- Detect continuous-wave signals using Hough transforms.
- Download open strain data, PSDs, and calibration envelopes.
- Extract configuration settings from PESummary files.
- Generate a
LALInferencerun script. - Launch a production-style parameter-estimation analysis.
This notebook demonstrates how large-scale Bayesian analyses are configured using the same software framework employed by the LVK Collaboration during O1–O3.
- Load posterior samples from a catalog of gravitational-wave detections.
- Define a population model for the primary black hole mass distribution.
- Infer hyperparameters such as the power-law slope
$\alpha$ and the minimum mass cutoff$m_{\min}$ . - Examine how the inferred parameters change when additional events are included.
- Explore the astrophysical implications of the resulting posterior distributions.
This notebook demonstrates how multiple detections can be combined to infer the underlying properties of the binary black hole population.
- What is the preferred value of the power-law slope
$\alpha$ governing the black hole mass spectrum? - What is the inferred lower mass cutoff
$m_{\min}$ ? - How would the inclusion of an event such as GW190814 affect the estimate of
$m_{\min}$ ? - How would an excess of low-mass black holes in future observing runs alter the inferred value of
$\alpha$ ?
- Population inference constrains the distribution of black hole masses rather than individual events.
- Additional detections can significantly shift the inferred hyperparameters.
- Events near the edge of the observed mass distribution strongly influence the estimate of
$m_{\min}$ . - A larger-than-expected number of low-mass black holes would favor a shallower mass spectrum.
- Generate Short Fourier Transforms (SFTs).
- Construct a time-frequency peakmap.
- Apply Doppler corrections for an assumed sky position.
- Compute the Frequency-Hough transform.
- Identify candidate peaks in the
$(f_0, \dot{f})$ plane. - Repeat the analysis with an incorrect sky position to study localization effects.
- Why do both positive and negative slopes appear in the Hough map?
- Why does the signal occupy multiple neighboring pixels around the true
$(f_0, \dot{f})$ values? - How does using an incorrect sky position affect the Doppler-corrected peakmap?
- Why is the Hough number count reduced when the source position is wrong?
- What happens to stationary instrumental lines after Doppler correction?
- The slope orientation depends on the ordering and sign convention of the spin-down grid.
- Finite frequency and spin-down resolution spreads the signal over several pixels.
- Correct sky localization sharpens the signal track and maximizes the Hough count.
- Incorrect Doppler correction smears the signal, reducing detection significance.
- Stationary noise lines are generally dispersed by Doppler correction, helping distinguish them from true astrophysical signals.
- Python
- NumPy
- Matplotlib
- Bilby
- LALSuite
- PESummary
- PyHough
After completing these extension topics, you will be able to:
- Configure production-style parameter-estimation analyses.
- Perform hierarchical inference on event populations.
- Understand how compact-object population models are constrained.
- Apply the Frequency-Hough method to continuous-wave searches.
- Explore advanced gravitational-wave analysis techniques used in current research.