Simulation-based inference with neural posterior estimation applied to X-ray spectral fitting - II. High-resolution spectroscopy with Athena X-IFU

Open Access

Table 1

Parameters true value and prior distribution used in the three models described in this paper.

`tbabs*(comptt+powerlaw)`
Component	Parameter	Truth	Prior
`tbabs`	n_H	0.1	$U$ $\mathcal{U}$ (0.01, 0.2)
`comptt`	z	0	Frozen
`comptt`	T₀	0.55	$U$ $\mathcal{U}$ (0.05, 2)
`comptt`	kT	2.5	$U$ $\mathcal{U}$ (2.1,4)
`comptt`	τ_p	2.5	$U$ $\mathcal{U}$ (0.5, 7)
`comptt`	approx	1	Frozen
`comptt`	norm	1	10^{$U$ (−1,1)}
`powerlaw`	α	2	$U$ $\mathcal{U}$ (0, 5)
`powerlaw`	norm	0.275	10^{$U$ (−1,1)}
`tbabs*relxillNS`
Component	Parameter	Truth	Prior
`tbabs`	n_H	0.2	$U$ $\mathcal{U}$ (0.1,0.3)
`relxillNS`	index₁	3	Frozen
`relxillNS`	index₂	3	Frozen
`relxillNS`	R_br	15	Frozen
`relxillNS`	a	0.2	$U$ $\mathcal{U}$ (0.05, 0.4)
`relxillNS`	i	30	Frozen
`relxillNS`	R_in	−1	Frozen
`relxillNS`	R_out	400	Frozen
`relxillNS`	z	0	Frozen
`relxillNS`	kT_BB	1.05	$U$ $\mathcal{U}$ (0.6, 2)
`relxillNS`	log ξ	3.1	$U$ $\mathcal{U}$ (1, 4.7)
`relxillNS`	A_Fe	2	Frozen
`relxillNS`	log N	16	Frozen
`relxillNS`	f	2.5	$U$ $\mathcal{U}$ (0.5, 4.5)
`relxillNS`	norm	0.01	10^{$U$ (−3,−1)}
`tbabs*(bapec+bapec)`
Component	Parameter	Truth	Prior
`tbabs`	nH	0.1	Frozen
`bapec`₁	kT₁	4	$U$ $\mathcal{U}$ (3,6)
`bapec`₁	Z₁	1.2	$U$ $\mathcal{U}$ (0.5, 2.5)
`bapec`₁	z₁	0.02	$U$ $\mathcal{U}$ (0, 0.05)
`bapec`₁	v₁	120	$U$ $\mathcal{U}$ (50, 200)
`bapec`₁	norm₁	1	10^{$U$ (−0.3,0.2)}
`bapec`₂	kT₂	3	$U$ $\mathcal{U}$ (1,4)
`bapec`₂	Z₂	1	$U$ $\mathcal{U}$ (0.5, 2.5)
`bapec`₂	z₂	0.02	$U$ $\mathcal{U}$ (0, 0.05)
`bapec`₂	v₂	100	$U$ $\mathcal{U}$ (50, 200)
`bapec`₂	norm₂	1.5	10^{$U$ (0,0.3)}

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.