Stellar flare detection in XMM-Newton with gradient-boosted trees

Home

All issues

Volume 708 (April 2026)

A&A, 708 (2026) A224

Full HTML

Open Access

Table A.1

Complete list of the features we used

-	Name	Description
1	N_BINS	Number of time bins
2	STDEV	Standard deviation
3	SKEW	Skewness
4	KURT	Kurtosis
5	AMPLIT	Amplitude (max(LC) − min(LC)/2
6	MEDABSDEV	MAD, Median deviation from median median(\|LC − median(LC)\|)
7	MEDMAXOFF	Max deviation from median on median max(\|LC − median(LC)\|)/\|median(LC)\|

8	CON_NSIGMA	Inverse survival function (constant model)
9	LIN_NSIGMA	Inverse survival function (linear model)
10	LIN_CON	Constant term (linear model)
11	LIN_CON_E^a	1σ error in LIN_CON
12	LIN_LIN	Linear term (linear model)
13	LIN_LIN_E^a	1σ error on LIN_LIN
14	QU_NSIGMA	Inverse survival function (quadratic model)
15	QU_CON	Constant term (quadratic model)
16	QU_CON_E^a	1σ error on QU_CON
17	QU_LIN	Linear term (quadratic model)
18	QU_LIN_E^a	1σ error on QU_LIN
19	QU_QU	Quadratic term (quadratic model)
20	QU_QU_E^a	1σ error on QU_QU
21	DIP1_NSIGMA	Inverse survival function (s.dip model)
22	DIP1_TMIN	Time of the minimum (s.dip model)
23	DIP1_TMIN_E^a	1σ error on DIP1_TMIN
24	DIP1_DT	Time duration of the dip (s.dip model)
25	DIP1_DT_E^a	1σ error on DIP1_DT
26	DIP1_FLMIN	Flux ad the minimum (s.dip model)
27	DIP1_FLMIN_E^a	1σ error on DIP1_FLMIN
28	DIP1_FLMAX	Constant term (s.dip model)
29	DIP1_FLMAX_E^a	1σ error on DIP1_FLMAX
30	DIP_NSIGMA	Inverse survival function (m.dip model)
31	DIP_TMIN	Phase of the minimum (m.dip model)
32	DIP_TMIN_E^a	1σ error on DIP_TMIN
33	DIP_DT	Phase duration of the dip (m.dip model)
34	DIP_DT_E^a	1σ error on DIP_DT
35	DIP_FLMIN	Flux ad the minimum (m.dip model)
36	DIP_FLMIN_E^a	1σ error on DIP_FLMIN
37	DIP_FLMAX	Constant term (m.dip model)
38	DIP_FLMAX_E^a	1σ error on DIP_FLMAX
39	DIP_PER	Time period (m.dip model)
40	DIP_PER_E^a	1σ error on DIP_PER
41	EC1_NSIGMA	Inverse survival function (s.ecl. model)
42	EC1_TIN	Begin time of eclipse (s.ecl. model)
43	EC1_TIN_E^a	1σ error on EC1_TIN
44	EC1_TFIN	End time of eclipse (s.ecl. model)
45	EC1_TFIN_E^a	1σ error on EC1_TFIN
46	EC1_FLMIN	Flux in the eclipse (s.ecl. model)
47	EC1_FLMIN_E^a	1σ error on EC1_FLMIN
48	EC1_FLMAX	Constant term (s.ecl. model)
49	EC1_FLMAX_E^a	1σ error on EC1_FLMAX
50	EC_NSIGMA	Inverse survival function (m.ecl. model)
51	EC_T1	Begin phase of eclipse (m.ecl. model)
52	EC_T1_E^a	1σ error on EC_T1
53	EC_T2	End phase of eclipse (m.ecl. model)
54	EC_T2_E^a	1σ error on EC_T2
55	EC_FL1	Flux level 1 (m.ecl. model)
56	EC_FL1_E^a	1σ error on EC_FL1
57	EC_FL2	Flux level 2 (m.ecl.model)
58	EC_FL2_E^a	1σ error on EC_FL2
59	EC_PER	Time period (m.ecl. model)
60	EC_PER_E^a	1σ error on EC_PER
61	FL_NSIGMA	Inverse survival function (flare model)
62	FL_CON	Constant term (flare model)
63	FL_CON_E^a	1σ error on FL_CON
64	FL_DFL	∆Flux at maximum (flare model)
65	FL_DFL_E^a	1σ error on FL_DFL
66	FL_TMAX	Time of maximum (flare model)
67	FL_TMAX_E^a	1σ error on FL_TMAX
68	FL_DT	Decay time (flare model)
69	FL_DT_E^a	1σ error on FL_DT
70	FL_RT	Rising time (flare model)
71	FL_RT_E^a	1σ error on FL_RT
72	F_NSIGMA_LINCON	f-test prob. in σs (linear-const.)
73	F_NSIGMA_QUCON	f-test prob. in σs (quadratic-const.)
74	F_NSIGMA_QULIN	f-test prob. in σs (quadratic-linear)
75	F_NSIGMA_DIP1CON	f-test prob. in σs (s.dip-const.)
76	F_NSIGMA_DIPCON	f-test prob. in σs (m.dip-const.)
77	F_NSIGMA_DIP1DIP	f-test prob. in σs (m.dip-s.dip)
78	F_NSIGMA_EC1CON	f-test prob. in σs (s.ecl.-const.)
79	F_NSIGMA_ECCON	f-test prob. in σs (m.ecl.-const.)
80	F_NSIGMA_EC1EC	f-test prob. in σs (m.ecl.-s.ecl.)
81	F_NSIGMA_FLCON	f-test prob. in σs (flare-const.)

82	TFRAC_MID20	%time between 0.9-1.1 of MEDIAN
83	TFRAC_BEL1S	%time below AVE-AVE_ERR^b
84	TFRAC_ABO1S	%time above AVE+AVE_ERR^b
85	TFRAC_BEL3S	%time below AVE-3AVE_ERR^b
86	TFRAC_ABO3S	%time above AVE+3AVE_ERR^b
87	TFRAC_BEL5S	%time below AVE-5AVE_ERR^b
88	TFRAC_ABO5S	%time frac. above AVE+5AVE_ERR^b
89	TFRAC_BEL1MAD	%time below MEDIAN-MAD^c
90	TFRAC_ABO1MAD	%time above MEDIAN+MAD^c
91	TFRAC_BEL3MAD	%time below MEDIAN-3MAD^c
92	TFRAC_ABO3MAD	%time above MEDIAN+3MAD^c
93	TFRAC_BEL5MAD	%time below MEDIAN-5MAD^c
94	TFRAC_ABO5MAD	%time above MEDIAN+5MAD^c
95	FLUX_P05	%flux at 5% of cumulative time
96	FLUX_P10	%flux below 10% of cumulative time
97	FLUX_P17.5	%flux below 17.5% of cumulative time
98	FLUX_P25	%flux below 25% of cumulative time
99	FLUX_P32.5	%flux below 32.5% of cumulative time
100	FLUX_P40	%flux below 40% of cumulative time
101	FLUX_P50	%flux below 50% of cumulative time
102	FLUX_P60	%flux below 60% of cumulative time
103	FLUX_P67.5	%flux below 76.5% of cumulative time
104	FLUX_P75	%flux below 75% of cumulative time
105	FLUX_P82.5	%flux below 82.5% of cumulative time
106	FLUX_P90	%flux below 90% of cumulative time
107	FLUX_P95	%flux below 95% of cumulative time
108	MEDIAN_01	Normalized MEDIAN

Notes. Complete list of the features we used with a brief explanation for each one and its formula, if necessary. We divided the list in three sublists separated by horizontal lines to show the model-independent statistical features, model-dependent features and CDF features, respectively.

^a _ERR has been written as _E to fit the table.

^b Weighted average and its 1σ error.

^c MEDIAN and MEDABSDEV.

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.