PPPSSR: PPPSSR

In SBCK: Statistical Bias Correction Kit

PPPSSR

Description

Apply the SSR transformation.

Details

Apply the SSR transformation. The SSR transformation replace the 0 by a random values between 0 and the minimal non zero value (the threshold). The inverse transform replace all values lower than the threshold by 0. The threshold used for inverse transform is given by the keyword 'isaved', which takes the value 'Y0' (reference in calibration period), or 'X0' (biased in calibration period), or 'X1' (biased in projection period)

Super class

SBCK::PrePostProcessing -> PPPSSR

Public fields

Xn

[vector] Threshold

Methods

Public methods

• PPPSSR$new()

• PPPSSR$transform()

• PPPSSR$itransform()

• PPPSSR$clone()

Inherited methods

• SBCK::PrePostProcessing$fit()
• SBCK::PrePostProcessing$predict()

---

Method new()

Create a new PPPSSR object.

Usage

PPPSSR$new(cols = NULL, isaved = "Y0", ...)

Arguments

cols

Columns to apply the SSR

isaved

Choose the threshold used for the inverse transform. Can be "Y0", "X0" and "X1".

...

Others arguments are passed to PrePostProcessing

Returns

A new 'PPPSSR' object.

---

Method transform()

Apply the SSR transform, i.e. all 0 are replaced by random values between 0 (excluded) and the minimal non zero value.

Usage

PPPSSR$transform(X)

Arguments

X

Data to transform

Returns

Xt a transformed matrix

---

Method itransform()

Apply the inverse SSR transform, i.e. all values lower than the threshold found in the transform function are replaced by 0.

Usage

PPPSSR$itransform(Xt)

Arguments

Xt

Data to transform

Returns

X a transformed matrix

---

Method clone()

The objects of this class are cloneable with this method.

Usage

PPPSSR$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Examples

## Start with data
XY = SBCK::dataset_like_tas_pr(2000)
X0 = XY$X0
X1 = XY$X1
Y0 = XY$Y0

## Define the PPP method
ppp = PPPSSR$new( bc_method = CDFt , cols = 2 )

## And now the correction
## Bias correction
ppp$fit(Y0,X0,X1)
Z = ppp$predict(X1,X0)

SBCK documentation built on Sept. 11, 2023, 5:10 p.m.