Draft Rubin Timeseries Features

	A	B	C	D	E	F	G	H	I	J	K	M
1		Feature	Number of floats	Description	Reference	Comments	Multiband or per-band?	compute on difference flux?	compute on total flux?	compute on forced difference flux?	compute on forced total flux?	Total number of floats
2
3	simple statistics	flux percentiles per band: 0, 5, 15.9, 25, 37.5, 50, 62.5, 75, 84.1, 95, 100	11				per-band			yes		11x6
4		template forced flux	1			to allow translation of difference flux percentiles to total flux.	per-band				yes	1x6
5		template validity range	2			assumes there are multiple templates in the timeseries
6		MAD	1				per-band			yes		1x6
7		skew	1				per-band			yes		1x6
8		robust kurtosis	1				per-band			yes		1x6
9		Stetson J	1				multiband?			yes
10		Stetson K	1				multiband?			yes
11		Normalized excess variance	1				per-band
12
13
14	periodic	linear term c	1				multiband				yes
15		f1, f2, f3	3			number of frequencies and harmonics chosen to be validated with cadence simulations	multiband				yes
16		Aij	12				multiband				yes
17		PHij	11				multiband				yes
18		freq_signif	1			need to consider non-white noise	multiband				yes
19		freq_signif_ratio_21	1				multiband				yes
20		freq_signif_ratio_31	1				multiband				yes
21		freq_varrat	1				multiband				yes
22		chisq/dof for periodic fit					multiband
23		other quality checks?
24		estimate of false alarm probability
25
26	stochastic	multi-band SF or DRW amplitude & uncertainty					multiband
27		multi-band SF or DRW timescale & uncertainty					multiband
28
29
30
31
32
33	transient	number of previous detections	1				per-band?
34		time of first detection
35		supernova/transient model parameters and uncertainties				Villar+ 2019 as implemented in Sanchez-Saez+ 21 (if computationally feasible)	per-band?
36		derived quantities?				e.g., peak time; representation of color
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100