MBCn {SBCK}

MBCn (Multivariate Bias Correction)

Description

Perform a multivariate bias correction.

Details

BC is performed with an alternance of rotation and univariate BC.

Public fields

n_features

[integer] Numbers of features

bc

[BC class] Univariate BC method

metric

[function] distance between two datasets

iter_slope

[Stopping class criteria] class used to test when stop

bc_params

[list] Parameters of bc

ortho_mat

[array] Array of orthogonal matrix

tips

[array] Array which contains the product of ortho and inverse of next

lbc

[list] list of BC method used.

Methods

Public methods

• MBCn$new()

• MBCn$fit()

• MBCn$predict()

• MBCn$clone()

Method new()

Create a new MBCn object.

Usage

MBCn$new(
  bc = QDM,
  metric = wasserstein,
  stopping_criteria = SlopeStoppingCriteria,
  stopping_criteria_params = list(minit = 20, maxit = 100, tol = 0.001),
  ...
)

Arguments

bc

[BC class] Univariate bias correction method

metric

[function] distance between two datasets

stopping_criteria

[Stopping class criteria] class use to test when to stop the iterations

stopping_criteria_params

[list] parameters passed to stopping_criteria class

...

[] Others arguments passed to bc.

Returns

A new 'MBCn' object.

Method fit()

Fit the bias correction method

Usage

MBCn$fit(Y0, X0, X1)

Arguments

Y0

[matrix: n_samples * n_features] Observations in calibration

X0

[matrix: n_samples * n_features] Model in calibration

X1

[matrix: n_samples * n_features] Model in projection

Returns

NULL

Method predict()

Predict the correction

Usage

MBCn$predict(X1, X0 = NULL)

Arguments

X1

[matrix: n_samples * n_features] Model in projection

X0

[matrix: n_samples * n_features or NULL] Model in calibration

Returns

[matrix or list] Return the matrix of correction of X1 if X0 is NULL, else return a list containing Z1 and Z0, the corrections of X1 and X0

Method clone()

The objects of this class are cloneable with this method.

Usage

MBCn$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

References

Cannon, A. J., Sobie, S. R., and Murdock, T. Q.: Bias correction of simulated precipitation by quantile mapping: how well do methods preserve relative changes in quantiles and extremes?, J. Climate, 28, 6938–6959, https://doi.org/10.1175/JCLI-D-14- 00754.1, 2015.

Examples

## Three bivariate random variables (rnorm and rexp are inverted between ref
## and bias)
XY = SBCK::dataset_gaussian_exp_2d(200)
X0 = XY$X0 ## Biased in calibration period
Y0 = XY$Y0 ## Reference in calibration period
X1 = XY$X1 ## Biased in projection period

## Bias correction
## Step 1 : construction of the class MBCn
mbcn = SBCK::MBCn$new() 
## Step 2 : Fit the bias correction model
mbcn$fit( Y0 , X0 , X1 )
## Step 3 : perform the bias correction, Z is a list containing
## corrections
Z = mbcn$predict(X1,X0) 
Z$Z0 ## Correction in calibration period
Z$Z1 ## Correction in projection period