Skip to Content

Hi,


SAP HANA SPS 12 was released yesterday 11th May 2016. If you want to delve deeper and see working examples of what’s new for predictive in SPS 12 then this blog can serve as your starting point.

Time series analysis is a major focus area in this release with Auto-ARIMA being the new algorithm on the block.


Auto-ARIMA expands the existing ARIMA capabilities of the Predictive Analysis Library (PAL) by automatically identifying the orders of an ARIMA (Seasonal-ARIMA-X) model according to information criterion such as AICC, AIC, and BIC. The orders can be represented as (p,d,q) (P,D,Q)m, where m is the seasonal period.


Auto-ARIMA is also available in the Application Functional Modeler (AFM) enabling WYSIWYG application development without the requirement to code in SQL Script.


Many additional enhancements have been made to other time series algorithms, including addition of custom confidence limits for singe, double, and triple exponential smoothing and new options for self organizing maps such as batch support.


Here are links to tutorials covering enhancements to the Application Function Modeler (AFM) for predictive:


AFM: 15. Getting Started with the Application Function Modeler

AFM: 16. PAL – Time Series – Auto ARIMA


Here are links to the new and enhanced SAP HANA predictive analysis library (PAL) algorithms:

Getting Started

PAL: 109. Getting Started with the Predictive Analysis Library

Time Series

PAL: 110. Time Series – Auto ARIMA Model

PAL: 111. Time Series – Auto ARIMA Predict

PAL: 112. Time Series – Single Exponential Smoothing Confidence

PAL: 113. Time Series – Double Exponential Smoothing Confidence

PAL: 114. Time Series – Triple Exponential Smoothing Confidence

PAL: 115. Time Series – Self Organizing Maps Parameters

Here’s the full playlist for what’s new: What’s New for Predictive SPS 12


Finally, if you’re interested to learn about other stuff that’s new in SPS 12 check out the following playlist:   SAP HANA SPS 12 – What’s New

Enjoy!

Philip


To report this post you need to login first.

Be the first to leave a comment

You must be Logged on to comment or reply to a post.

Leave a Reply