Equation of state inference
We released an article on cold dense matter equation of state inference based on astrophysical and experimental constraints and showed the impact individual observations have on the current constraints and show that the priors on NS properties and EoS framework plays a bigger role in defining the inferred EoS. [ Link to the open access paper]
The impact is quite stark. Individual observations turn out to have small effect given we have 6-7 broadly constraining observations (GW170817, M-R from NICER, high mass pulsar observations, and PREX constraint). The movement in the M-R changes by small amounts when these are individually removed, hence implying unless we have a strong new constraint, the gains in EoS constraints will be minimal.
On the other hand, EoS framework plays a huge role! EoS framework determines what types of EoSs can be probed during Bayesian inference. We swap the traditionally used (Gamma) spectral representation with an improved (Upsilon) causal spectral representation which ensures sound speed causality and continuity. This new framework, Upsilon, v the previous, Gamma, predict substantially distinct EoS ranges. Upsilon spectral prefers a substantially backward bending EoSs and attains an EoS which has 2.3 times higher likelihood than any attained by Gamma spectral representation.
Additionally, NS intrinsic parameter priors also play a substantial role in this inference. We applied three different intrinsic parameter priors based on different astrophysical expectations and show that the attained posterior is shifted by about 1km on either direction based on these priors. This effect is larger than that from removing individual observations.
For this we enhanced RIFT's capabilities to beyond inferring gravitational wave source's intrinsic and extrinsic parameters. It now has capability to perform parameter estimation for any types of properties as long as there is a likelihood evaluation prescribed by the user, and we call this feature Hyperposterior Pipeline - HyperPipe. User manual at: RIFT-HyperPipe.
A. Kedia, R. O'Shaughnessy, L. Wade, A. Yelikar, Exploring hidden priors when interpreting gravitational wave and electromagnetic probes of the nuclear equation of state, arXiv:2405.17326 [astro-ph.HE] (2024).
The impact is quite stark. Individual observations turn out to have small effect given we have 6-7 broadly constraining observations (GW170817, M-R from NICER, high mass pulsar observations, and PREX constraint). The movement in the M-R changes by small amounts when these are individually removed, hence implying unless we have a strong new constraint, the gains in EoS constraints will be minimal.
On the other hand, EoS framework plays a huge role! EoS framework determines what types of EoSs can be probed during Bayesian inference. We swap the traditionally used (Gamma) spectral representation with an improved (Upsilon) causal spectral representation which ensures sound speed causality and continuity. This new framework, Upsilon, v the previous, Gamma, predict substantially distinct EoS ranges. Upsilon spectral prefers a substantially backward bending EoSs and attains an EoS which has 2.3 times higher likelihood than any attained by Gamma spectral representation.
Additionally, NS intrinsic parameter priors also play a substantial role in this inference. We applied three different intrinsic parameter priors based on different astrophysical expectations and show that the attained posterior is shifted by about 1km on either direction based on these priors. This effect is larger than that from removing individual observations.
For this we enhanced RIFT's capabilities to beyond inferring gravitational wave source's intrinsic and extrinsic parameters. It now has capability to perform parameter estimation for any types of properties as long as there is a likelihood evaluation prescribed by the user, and we call this feature Hyperposterior Pipeline - HyperPipe. User manual at: RIFT-HyperPipe.
A. Kedia, R. O'Shaughnessy, L. Wade, A. Yelikar, Exploring hidden priors when interpreting gravitational wave and electromagnetic probes of the nuclear equation of state, arXiv:2405.17326 [astro-ph.HE] (2024).