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.

<strong>PAL</strong>_KORD_CONS_ITEMS_TBL<br />

3.5 Time Series Algorithms<br />

Financial market data or economic data usually comes with time stamps. Predicting the future values, such as<br />

stock value for tomorrow, is of great interest in many business scenarios. Quantity over time is called time<br />

series, and predicting the future value based on existing time series is also known as forecasting. In this<br />

release of <strong>PAL</strong>, three smoothing based time series models are implemented. These models can be used to<br />

smooth the existing time series and forecast. In the time series algorithms, let x t be the observed values for the<br />

t-th time period, and T be the total number of time periods.<br />

3.5.1 ARIMA<br />

The auto regressive integrated moving average (ARIMA) algorithm is famous in econometrics, statistics and<br />

time series analysis. An ARIMA model can be written as ARIMA (p, d, q), where p refers to the auto regressive<br />

order, d refers to integrated order, and q refers to the moving average order. This algorithm helps you<br />

understand the time series data better and predict future data in the series.<br />

The auto regressive integrated moving average with intervention (ARIMAX) algorithm is an extension for<br />

ARIMA. Compared with ARIMA, an ARIMAX model can not only get the internal relationship with former data,<br />

but also take external factor into consideration.<br />

320 P U B L I C<br />

<strong>SAP</strong> <strong>HANA</strong> <strong>Predictive</strong> <strong>Analysis</strong> <strong>Library</strong> (<strong>PAL</strong>)<br />

<strong>PAL</strong> Functions

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

Saved successfully!

Ooh no, something went wrong!