Originally posted in Spanish at https://sasybi.blogspot.com.es/2015/07/sas-visual-a...
SAS Visual Analytics is a business analytics solution that allows you to visually explore all data in an easy-to-use platform that's accessible to users of all levels without statistical, technical, or design skills.
Visual Analytics within BI solutions that are available on the market are positioned within the analytical displays solutions group. In this group, we have solutions such as QlikView, Tableau, and TIBCO Spotfire, amongst others. In summary the proposed solutions have:
SAS Visual Analytics offers a complete analytical platform for displaying information, allowing you to identify patterns and relationships in data that were not previously apparent. The interactive capabilities of self-service BI and reporting combine with advanced analytics for all to help discover knowledge of data of any size and type.
Let us now look at the features of the tool, analyzing each of the main modules, and its technical architecture:
SAS Visual Analytics has a module for importing data and other data preparation based on SQL which allows adapting imported data to the optimal structure for its exploitation. For most potential analyses, the recommended tool works on a table that consolidates aggregate information from multiple tables and starting file. This is the classic board obtained as N junction fact tables and dimensions. The tool also enables the option of working with a model in which star felling facts and dimensions would be separate tables.
The tool has a module for data preparation that allows data transformation on imported data for performance analysis based on a SQL query builder. This module may, thus, stop a little when transformations to be performed are fairly complex. In this case, I propose using SAS Enterprise Guide, offering the choice of Visual Analytic Pro (Visual Analytics + Enterprise Guide).
With the fields of the imported tables, it is relatively easy to derive the calculated fields using elements in a simple way, giving access to a powerful expression editor.
One of the main differences of SAS over other analytical tools are its display analytic capabilities (predictive techniques, time series, associations, etc.) based on the long experience of SAS tools such as SAS Enterprise Miner. The algorithms apply predictive analytics for automatic detection, and you can get detailed info on the selected algorithm. You can easily create decision trees for groups or classifications in the data, as well as box-plot diagrams to learn more about the distribution of data.
The ability to easily obtain time series for process Forecasts. These processes are very simple to implement, but would fall short if we think of a more industrialized forecast that would make a massive entry which would result forecast for other systems (e.g. forecast need for stocks), in these cases it is advisable to go solutions SAS Forecast Server type.
In predictive processes, we can use the functionality " underlying factors "that allows us to evaluate how other variables affect our prediction can perform scenario analysis and simulations, "what-if".
It has the ability to connect through add-in to Visual Statistics for processes that need more advanced statistical analysis.
Utilities to learn about the relationships between variables, such as correlation matrices. Descriptive statistics that provide insight into the distribution of values in the variables (minimum, maximum, average, zero, etc.)
Report Designer very intuitive use (drag and drop). We can easily create reports or dashboard using the graphics and visualization objects as include indicators or classifiers select.
Ability to incorporate dashboards analysis documents obtained in the process of exploratory analysis.
Once you designed a serial graphic objects on a document we can define interactions between them, to relate the selections made some of them to other objects or to define navigation between them.
SAS Visual Analytics incorporates multiple possible visualization box plots, heat maps, animated bubble charts, network diagrams, decision trees, geolocation. Likewise, auto charting capabilities help determine the most appropriate graph to display the data according to the elements selected for analysis. A bar overview allows you to zoom on the range of data that you want, without losing the whole picture.
Dimensions and hierarchies Organization for OLAP analysis multidimensional.
Creation, display, publication and distribution of multi-device analysis and reporting. Integration with Office Outlook, SharePoint, Excel and Power Point
Response times are nimble because the data is loaded into memory based on SAS LASR (server analytical high performance memory). It also has solution oriented Cloud with an on-premise option.
In short it is a powerful analytical tool display, which is an interesting option for companies without having to make a large initial investment, want to start making analytical, with the ability to scale and grow into other tools.