Subscribe to DSC Newsletter

Difference between Reporting (MIS),Analytics and Advanced Analytics

Hi all!,

 

As the discussion title says I would like to understand the difference between Reporting, Analytics and Advanced Analytics.I get a feeling that these terms are used interchangeably by most of the Analytics folks themselves (especially in my part of the world) :).

 

Regards,

Sharath

Views: 10820

Reply to This

Replies to This Discussion

reporting (MIS) is all about what has happened in the past or currently happening. dashboards, charts, graphs, etc

Analytics is finding out the reasons of what these trends are occuring - some slice & dice, data mining

Advance Analytics is all about predicting what will happen next & doing optimization on what will happen best !!

 

When data under analysis is known the processing is called Descriptive Analytics. Known data is the recorded history. There are many techniques to do this type of analytics but besides of graphic depiction, filtering, decomposition, and restructuring all others are domain specific. From accounting a well known example is the du Pont analysis. In a nutshell Descriptive Analytics is about explaining why something happened and is seen by many people as a reporting tool.

 

When data under analysis is unknown or partially unknown the processing is called Predictive Analytics. Data could be unknown because it is lost history or hasn’t happened yet. The main techniques are extrapolation and what if analysis. Most people view Predictive Analytics as the true form of analytics. In a nutshell this form of analytics uses available data and makes assumptions to predict the missing data. This set of techniques brings very good results is the unknown data is close to the known data. It could also be a very complex processing when the number of variables is too large (curse of dimensionality).

 

When data under analysis is unknowable there are several techniques of inferring properties like statistics, limits, averages, etc. My company uses Heuristic Analytics which employs widely available data and a mathematical model (observer) to predict the unknowable data. You could find more at http://www.heuristicanalytics.com/pdf/ITCknow.pdf. In a nutshell Heuristic Analytics looks somewhere else to find supporting data. Since between unknown and unknowable it is a very fine line the Heuristic Analytics could be used when the unknown data is too far from the known to allow reasonable extrapolations.

 

RSS

On Data Science Central

© 2019   AnalyticBridge.com is a subsidiary and dedicated channel of Data Science Central LLC   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service