SAS to help Sebi implement better compliance mechanism

The market watchdog Sebi today roped in leading business analytics software services provider SAS to help improve its investigation department and put in place a robust surveillance mechanism to ensure unbiased trading platforms.

The Securities and Exchange Board of India (Sebi) intends to build analytical models using SAS EnterpriseMiner to identify known market manipulation patterns like circular trading, pump and dump, insider trading and front-running, SAS said in a release here.
"Pulling data into our warehouse and analyzing it the next day is our single most important task," Sebi general manager Avneesh Pandey said in the release.
"Data loading is crucial for us. We collect about 25GB of data per day and project it reaching 80GB per day in two years. With SAS help, we can accelerate the process of utilising this data," he said.
Increasing regulations and scrutiny demand sophisticated analysis and monitoring to spot market malpractices. SAS would help speed up investigation of suspicious transactions and boost investor confidence, SAS said in the statement.
With SAS, we maintain uniformity, making data easier to analyse. The result is smarter investigations for quicker results," Pandey, who heads technology projects at Sebi's surveillance, investigations and research departments, said.
SAS Institute regional director for Southeast Asia and chief executive and managing director Sudipta K Sen said, "SAS with its technical acumen in business analytics and understanding of the Indian market will be able to provide Sebi a single-view of customers across the exchanges."
To handle data growth and build a tighter fraud surveillance and investigation platform, Sebi wants a robust warehouse, high-end analytics, and better predictive modeling and text mining. Besides addressing growing volumes of data, SAS gives Sebi a single view of customers across exchanges.
The market regulator can now establish relationships between market participants and generate more accurate fraud alerts based on participants' behavior.