Output files
Once you finished a run with pyaneti, the code creates a series of output files stored in outpy/star_out (where star is the name of your modelled system). These files consist of the posterior distributions, ready-to-publish plots, residuals, etc. Here you will find a brief description of each file:
This file shows first the information about the MCMC run:
--------------------------------------------------------------
Summary:
N_chains = Number of chains in the MCMC
N_iter = Number of iterations saved in the posterior distribution file
thin_factor = Thin factor
N_rv_data = Number of RV data points
N_tr_data = Number of light curve data points
N_data = Total number of data points (RV + light curve)
N_pars = Number of fitted parameters
chi2_rv = Chi^2 for RV fit
chi2_tr = Chi^2 for transit fit
chi2 = Total Chi^2
dof = Degrees of Freedom
chi2/dof = Reduced Chi^2
ln likelihood_rv = Ln(Likelihood) for RV fit
ln likelihood_tr = Ln(Likelihood) for transit fit
ln likelihood = Ln(Likelihood) for total fit
BIC = Bayesian Information Criteria (BIC)
AIC = Aikaike Information Criteria(AIC)
After this, you will find your input stellar parameters (stellar mass, radius, and effective temperature, you have to specify these values in the input_fit.py).
--------------------------------------------------------------
INPUT STELLAR PARAMETERS
--------------------------------------------------------------
M_* = mean_value - error_bar + error_bar solar masses
R_* = mean_value - error_bar + error_bar solar radii
T_* = mean_value - error_bar + error_bar K
The next information displayed is the fitted and derived parameters for each planet. Each planet is labeled as star n, where star is your fitted system name ad n follows the alphabetical sequence [b,c,d,...]. The inferred parameter value and its uncertainty are given by the median and 68.3% credible interval of the posterior distribution. Note that the code will only show RV parameters if you are not fitting transits, or vice-versa.
--------------------------------------------------------------
Parameters star n
-------------------------Fitted-------------------------------
T0 = Time of inferior conjunction of the planet days
P = Orbital period days
e = Orbit eccentricity
w = Angle of periastron of the star deg
b = Impact parameter
a/R* = Scaled semi-major axis
rp/R* = Scaled planet radius
K = Induced Doppler semi-amplitude m/s
-------------------------Derived------------------------------
Mp = Planet mass (units depends on the units specified in the input file)
Rp = Planet radius (units depends on the units specified in the input file)
Tperi = Time of periastron days
i = Orbit inclination deg
a = Orbit semi-major axis AU
Insolation = Planet insolation F_Earth
rho* = Stellar density coming from the transit fit (transit)
rho* = Stellar density coming from the input stellar parameters (stellar paramters)
rho_p = Planet density g/cm^3
g_p = Planet's surface gravity from fitted parameters cm/s^2 (K and Rp/R*, see Southworth et al., 2007, MNRAS, 379, L11)
g_p = Planet's surface gravity from inferred planet mass and radius cm/s^2 (planet parameters)
Teq = Equilibrium temperature assuming albedo = 0 K (albedo=0)
T_tot = Total eclipse duration (from contact points 1 to 4) hours
T_full = Full eclipse duration (from contact points 2 to 3) hours
--------------------------------------------------------------
The file ends showing the extra fitted and derived parameters, such as limb darkening coefficients, systemic velocities and jitter terms.
-------------------- Other parameters -----------------------
q1 = Parametrization of quadratic limb darkening coefficients (see Kipping 2013)
q2 = Parametrization of quadratic limb darkening coefficients (see Kipping 2013)
u1 = Linear term of limb darkening coefficient of quadratic parametrization
u2 = Quadratic term of limb darkening coefficient of quadratic parametrization
Sys. vel. INSTRUMENT = Systemic velocity for each instrument (one line for instrument) m/s
--------------------------------------------------------------
HARPS-N jitter = Jitter term for each RV instrument (one line for instrument) m/s
tr jitter = Jitter term for light curve
--------------------------------------------------------------
This file contains the fitted and derived parameters of the system in a \newcommand LaTeX environment. Each parameter is given as
\newcommand{\commandname}[1][units]{median _{-error_bar}^{ + error_bar }$~#1}
These commands are really useful when writing papers, thesis, etc.
This is a CSV file containing the parameter samples used to create the posterior distributions. Each column corresponds to a different parameter and each line to a different chain at different iterations. The columns are organized as follows:
| Column number | Parameter |
|---|---|
| 1 | Number of saved iteration |
| 2 | log Likelihood for the current line |
| 3 | Chi^2 for RV data for the current line |
| 4 | Chi^2 for light curve data for the current line |
The rest of the columns contain planet, orbital, and Gaussian Processes parameters. The file contains a header indicating the parameter that each column describes.
Note: This file may use a high amount of computer space for multi-planet systems.
This file contains the residuals of the RV data after extracting the inferred model. This file may be useful to search for the extra signals not accounted for in the fitted model.
Note: This file is not created if you are not fitting RV data.
This file contains your light curve input file after a 10-sigma clipping algorithm.
If you used the textra variable in your input file, the timestamps of star_new_lc.dat are the original timestamps of your light curve file + textra.
The file star_tango_input.py contains an input file that you can use to animate your transits with tango.
You only need to copy the star_tango_input.py as an input.py inside a target directory inside tango together with the light curve file corresponding to a given system.
Note:
- The
star_tango_input.pyassumes that you will animate thestar_new_lc.datfile. If you desire to animate another light curve for a given system, you need to change thelcnameparameter in the input file. - For more details on how to use
tangoclick here. - This file is not created if you do not fit transits.
This file corresponds to the invidual induced RV signal for each planet b, c, ..., after the substraction of the other signals (such as other planets or trends). The inferred model is plotted as a thick black line and the instruments are labelled as different colours/symbols.
This is an example of a star[b..]_rv file
Note: This file is only created for the planets with fitted RV signal.
This file corresponds to the invidual transit signal for each planet b, c, ..., after the substraction of the other planets. The inferred model is plotted as a thick black line and the solid circles are data.
This is an example of a star[b..]_tr file
Note: This file is only created for the planets with fitted transit signal.
This file corresponds to a plot of whole light curve data together with the inferred model of all the transiting planets.
Note: This file is created if there is at least one fitted transiting planet.
This file corresponds to a plot of whole RV data set together with the inferred model of all the RV signals. The RV measurements are plotted with the systemic velocities subtracted.
Note: This file is created if there is at least one fitted RV signal.
This file contains plots of histograms for each fitted parameters (the parameters contained in star_all_data.dat). Each square in the file corresponds to the parameter labelled in its bottom.
Blue shaded region corresponds to the marginal posterior distribution for each parameter. Green shaded region corresponds to the prior distribution for each parameter. Solid red line marks the median value of the distribution; dashed red lines limit the 68.3% of the credible interval. Yellow line marks the mode of the distribution.
This kind of plots of posterior and prior together are useful to analise visually the role of the priors on the final parameter estimation.
This file contains correlation plots between fitted parameters (the parameters contained in star_all_data.dat). Each square in the file corresponds to the parameter labelled in its bottom.
This file contains a corner plot of the fitted parameters (the parameters contained in star_all_data.dat).
Note: You have to have corner package in your machine. This file is created only if you include is_corner_plot = True in your input file. This option also cancels star_posterior.pdf and star_Correlations.pdf plots.
