Table 2
List of light-curve features.
| Feature name | Descriptiona | Referenceb |
|
|
||
| intercept (slope) | Intercept (slope) of a linear regression to the light curve | D07 |
| A ij | Amplitude of the ith frequency and jth harmonic | D07 |
| φ ij | Phase of the ith frequency and jth harmonic | D07 |
| f i | ith frequency obtained from GLS | D07 |
| P i | Peak in the GLS periodogram of the ith frequency | D07 |
| vari | Variance left after ith fit of Fourier model | D07 |
| msei | Mean squared error of ith fit of Fourier model | D07 |
| skew | Skewness of y | R11 |
| small_kurtosis | Small sample kurtosis of y | R11 |
| std | Standard deviation of y | R11 |
| max_slope | max{ (yi + 1−yi)/(ti + 1−ti) } | R11 |
| amplitude | max(y)−min(y) | R11 |
| median_absolute_deviation | Median absolute deviation (MAD) of y | R11 |
| median_buffer_range_percentage | Fraction of points in { y } with amplitude within <0.1 of median(y) | R11 |
| pair_slope_trend | For the set  the ratio N+/N− |
R11 |
| flux_percentile_ratio_mid_k | F 50−k/ 2,50 + k/ 2 / F 5,95 | R11 |
| percent_amplitude | max(| F−median(F) | )/median(F) | R11 |
| percent_difference_flux_percentile | F5,95/ median(F) | R11 |
| freq_amplitude_ratio_21 (31) | Amplitude ratio of 2nd (3rd) to 1st component of the Fourier model | R12 |
| freq_frequency_ratio_21 (31) | Frequency ratio of 2nd (3rd) to 1st component of the Fourier model | R12 |
| freq_model_max(min)_delta_mags | Difference in magnitudes between the two maxima (minima) of y2P | R12 |
| freq_model_phi1_phi2 | (φmin,1−φmax,1)/(φmin,1−φmax,2) (for ym,2P) | R12 |
| freq_rrd | Boolean that is 1 if freq_frequency_ratio_21 (or 31) is within 0.0035 of 0.746 | R12 |
| gskew | (median(y)−median(y0)) + (median(y)−median(y1−p)) with p = 0.03 | R12 |
| scatter_res_raw | MAD(y−ym)/MAD(y) | D11 |
| p2p_scatter_2praw |
|
D11 |
| p2p_scatter_over_mad |
|
D11 |
| p2p_scatter_pfold_over_mad |
|
D11 |
| medperc90_2p_p | 90th percentile of { y−ym,2P }/90th percentile of { y−ym,P } | D11 |
| fold2P_slope_10percentile (90) | 10th (90th) percentile of slopes y2P | R12 |
| R1c | (φmax,1−φmin,1)/(φmin,1−φmax,2) (for ys,2P) | R12, this work |
| R2 | (ys,2P(φmax,1)−ys,2P(φmin,1))/(ys,2P(φmin,1)−ys,2P(φmax,1)) | This work |
Notes.
We use the following notation: the light-curve magnitudes at times ti are denoted by y(ti) or yi, the magnitudes phased with period P at phase φi as yP(φ), the harmonic (Fourier) model as ym, the smooth spline mode as ys. φmax(min),i denotes the phase corresponding to the ith maximum (minimum) of a phased curve, y(φmax(min),i) the corresponding value. N+ and N− denote the number of positive and negative members of a set, respectively. Fa,b is the difference in flux between the percentile a and b of the fluxes implied by y. ya:b are the subset of y whose members lie between the ath and bth percentile;
D07 = Debosscher et al. (2007); R11 = Richards et al. (2011); R12 = Richards et al. (2012a); D11 = Dubath et al. (2011);
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.