Data Mining Algorithms In R/Packages/CCMtools/DI
Description
[edit | edit source]This function calculates the Daily Intensity (DI) cost function for precipitation. This error is supposed to be due to a misclassification (i.e., the elements should be classified according to sequence cl.check, while they are in practice allocated to the clusters according to the sequence cl.proj).
Usage
[edit | edit source]DI(cl.calibration, cl.check, cl.proj, DataS.Calibration, DataS.check)
Arguments
[edit | edit source]- cl.calibration Sequence of clusters for time period 1 (TP1)
- cl.check Reference sequence (i.e., numerical vector) of clusters (i.e., this is the sequence we should have) for time period 2 (TP2).
- cl.proj Sequence of clusters in practice for TP2.
- DataS.Calibration Precipitation data time series for TP1.
- DataS.check Precipitation data time series for TP2.
Details
[edit | edit source]For details about these cost functions, see the reference below. M. Vrac, P. Yiou. "Weather regimes designed for local precipitation modelling: Application to the Mediterranean bassin". Submitted, JGR-A, 2009
Value
[edit | edit source]Returns a list with arguments:
- DI Daily Intensity (DI) error costs, depending on station s and day d.
- MDI Means DI costs, depending on station s.
- RMDI Regional MDI providing a global (spatial and temporal) view of the DI error
cost.
Author
[edit | edit source]M. Vrac (mathieu.vrac@lsce.ipsl.fr))