25.10.2016 Views

SAP HANA Predictive Analysis Library (PAL)

sap_hana_predictive_analysis_library_pal_en

sap_hana_predictive_analysis_library_pal_en

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

3.5.10 Triple Exponential Smoothing<br />

Triple exponential smoothing is used to handle the time series data containing a seasonal component. This<br />

method is based on three smoothing equations: stationary component, trend, and seasonal. Both seasonal<br />

and trend can be additive or multiplicative. <strong>PAL</strong> supports multiplicative triple exponential smoothing and<br />

additive triple exponential smoothing. For additive triple exponential smoothing, an additive damped method is<br />

also supported.<br />

Multiplicative triple exponential smoothing is given by the below formula:<br />

Additive triple exponential smoothing is given by the below formula:<br />

S t = α × (X t − C t−L ) + (1 − α) × (S t−1 + B t−1 )<br />

B t = β × (S t − S t−1 ) + (1 − β) × B t−1<br />

C t = γ × (X t − S t ) + (1 − γ) × C t−L<br />

F t+m = S t + m × B t + C t−L+1+((m−1)mod L)<br />

The additive damped method of additive triple exponential smoothing is given by the below formula:<br />

S t = α × (X t − C t−L ) + (1 − α) × (S t−1 + Φ × B t−1 )<br />

B t = β × (S t − S t−1 ) + (1 − β) × Φ × B t−1<br />

C t = γ × (X t − S t ) + (1 − γ) × C t−L<br />

F t+m = S t + (Φ + Φ 2 + ... + Φ m ) × B t + C t−L+1+((m−1)mod L)<br />

Where:<br />

Table 290:<br />

α Data smoothing factor. The range is 0 < α

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!