CONAN._classes.load_rvs.get_decorr

CONAN._classes.load_rvs.get_decorr#

CONAN._classes.load_rvs.get_decorr(T_0=None, Period=None, K=None, Eccentricity=None, omega=None, sesinw=None, secosw=None, gamma=0, delta_BIC=-5, decorr_bound=(-1000, 1000), exclude_cols=[], enforce_pars=[], exclude_pars=[], show_steps=False, plot_model=True, use_jitter_est=False, setup_baseline=True, setup_planet=False, custom_RVfunc=None, verbose=True)#

Function to obtain best decorrelation parameters for each rv file using the forward selection method. It compares a model with only an offset to a polynomial model constructed with the other columns of the data. It uses columns 0,3,4,5 to construct the polynomial trend model. The temporary decorr parameters are labelled Ai,Bi for 1st & 2nd order in column i. Decorrelation parameters that reduces the BIC by 5(i.e delta_BIC = -5, 12X more probable) are iteratively selected. The result can then be used to populate the rv_baseline() method, if use_result is set to True.

Parameters:

  • T_0 (floats, None;) – RV parameters of the planet. T_0 and P must be in same units as the time axis (cols0) in the data file. if float/int, the values are held fixed. tuple/list of len 2 implies gaussian prior as (mean,std) while len 3 implies [min,start_val,max]

  • Period (floats, None;) – RV parameters of the planet. T_0 and P must be in same units as the time axis (cols0) in the data file. if float/int, the values are held fixed. tuple/list of len 2 implies gaussian prior as (mean,std) while len 3 implies [min,start_val,max]

  • K (floats, None;) – RV parameters of the planet. T_0 and P must be in same units as the time axis (cols0) in the data file. if float/int, the values are held fixed. tuple/list of len 2 implies gaussian prior as (mean,std) while len 3 implies [min,start_val,max]

  • Eccentricity/sesinw (floats, None;) – RV parameters of the planet. T_0 and P must be in same units as the time axis (cols0) in the data file. if float/int, the values are held fixed. tuple/list of len 2 implies gaussian prior as (mean,std) while len 3 implies [min,start_val,max]

  • omega/secosw (floats, None;) – RV parameters of the planet. T_0 and P must be in same units as the time axis (cols0) in the data file. if float/int, the values are held fixed. tuple/list of len 2 implies gaussian prior as (mean,std) while len 3 implies [min,start_val,max]

  • delta_BIC (float (negative);) – BIC improvement a parameter needs to provide in order to be considered relevant for decorrelation. Default is conservative and set to -5 i.e, parameters needs to lower the BIC by 5 to be included as decorrelation parameter.

  • decorr_bound (tuple of size 2;) – bounds when fitting decorrelation parameters. Default is (-1000,1000)

  • exclude_cols (list of int;) – list of column numbers (e.g. [3,4]) to exclude from decorrelation. Default is [].

  • enforce_pars (list of int;) – list of decorr params (e.g. [‘B3’, ‘A5’]) to enforce in decorrelation. Default is [].

  • exclude_pars (list of str;) – list of decorr parameters (e.g. [‘B3’, ‘A5’]) to exclude from decorrelation. Default is [].

  • show_steps (Bool, optional;) – Whether to show the steps of the forward selection of decorr parameters. Default is False

  • plot_model (Bool, optional;) – Whether to plot data and suggested trend model. Defaults to True.

  • use_jitter_est (Bool, optional;) – Whether to use the jitter estimate to setup the baseline model. Default is False.

  • setup_baseline (Bool, optional;) – whether to use result to setup the baseline model. Default is True.

  • setup_planet (Bool, optional;) – whether to use input to setup the planet parameters. Default is False.

  • verbose (Bool, optional;) – Whether to show the table of baseline model obtained. Defaults to True.

_rvdecorr_result#

list of decorr result for each rv.

Type:

list;

_rvmodel#

list to hold determined trendmodel for each rv.

Type:

list;

_rv_pars#

dictionary of RV parameters

Type:

dict;

Returns:

decorr_result – list containing result object for each lc.

Return type:

list of result object

Contents