Correcting systematics

This sub-package defines classes which help remove instrument systematics or variability from time series photometry data.

Classes provided by this package should inherit from an abstract Corrector class, which provides three key methods:

    .correct(**options) -> Returns a systematics-corrected LightCurve.
    .diagnose(**options) -> Returns figures which elucidate the correction.
    .interact() -> Returns a widget to tune the options interactively (optional).

Classes currently provided are KeplerCBVCorrector, SFFCorrector, and PLDCorrector.


DesignMatrix(df[, columns, name, prior_mu, …])

A matrix of column vectors for use in linear regression.


A set of design matrices.

KeplerCBVCorrector(lc[, cbv_array, …])

Remove systematic trends from Kepler light curves by fitting Cotrending Basis Vectors (CBVs).


Implements the Pixel Level Decorrelation (PLD) systematics removal method.


Remove noise using linear regression against a DesignMatrix.


Special case of RegressionCorrector where the DesignMatrix includes the target’s centroid positions.

TessPLDCorrector(tpf[, aperture_mask])

Correct TESS light curves by detrending against local pixel time series.

Class Inheritance Diagram

Inheritance diagram of lightkurve.correctors.designmatrix.DesignMatrix, lightkurve.correctors.designmatrix.DesignMatrixCollection, lightkurve.correctors.cbvcorrector.KeplerCBVCorrector, lightkurve.correctors.pldcorrector.PLDCorrector, lightkurve.correctors.regressioncorrector.RegressionCorrector, lightkurve.correctors.sffcorrector.SFFCorrector, lightkurve.correctors.tesspldcorrector.TessPLDCorrector