OSCAR Student Jad Rayes Applyies Large Scale Data Driven Techniques to Insider Trading
Insider trading is something that we’ve heard about time and
again in the news. It made headlines back in 2001 when Enron executives were
charged with insider trading and accounting fraud. It has been a significant
challenge for the financial regulatory body of the U.S government, the SEC, to
both prevent and identify. I became fixated on applying data mining techniques
to aid financial regulators in identifying and preventing illegal insider
trading. I feel that using recent advances in big data to make the market safer
and fairer, is both a noble and important pursuit. I know that I want to pursue
a career as a data scientist, and being able to explore and invent techniques
to solve problems with big data is crucial. It is also important to my future
career goals that I am able to use my technical background and skill set to
solve problems of vastly different types, in different domains. My work this
summer is a continuation of the work I have been doing since September of last
year. I am working with insider transaction data from about 400 different
companies in different sectors. Each week I make it a goal to read a paper
related to my project. I write new scripts to perform different analysis on my
datasets; and will often experiment with different mathematical techniques to
summarize and describe new trends and correlations I observe. I spend a fair
amount of time using different python packages to both build creative plots and
figures, and summarize my results concisely. I find that being able to explain
your results clearly is impossible without the use of a good graphing tool.
Since my project began, I have discovered how information propagates between
insiders, how pricing data on stock is affected by financial news, and how
certain network topologies are useful for representing relationships between
insiders. The most important resources I have discovered are the faculty and
graduate students who have taught me how to form a hypothesis and conduct
research.
Insider trading is something that we’ve heard about time and
again in the news. It made headlines back in 2001 when Enron executives were
charged with insider trading and accounting fraud. It has been a significant
challenge for the financial regulatory body of the U.S government, the SEC, to
both prevent and identify. I became fixated on applying data mining techniques
to aid financial regulators in identifying and preventing illegal insider
trading. I feel that using recent advances in big data to make the market safer
and fairer, is both a noble and important pursuit. I know that I want to pursue
a career as a data scientist, and being able to explore and invent techniques
to solve problems with big data is crucial. It is also important to my future
career goals that I am able to use my technical background and skill set to
solve problems of vastly different types, in different domains. My work this
summer is a continuation of the work I have been doing since September of last
year. I am working with insider transaction data from about 400 different
companies in different sectors. Each week I make it a goal to read a paper
related to my project. I write new scripts to perform different analysis on my
datasets; and will often experiment with different mathematical techniques to
summarize and describe new trends and correlations I observe. I spend a fair
amount of time using different python packages to both build creative plots and
figures, and summarize my results concisely. I find that being able to explain
your results clearly is impossible without the use of a good graphing tool.
Since my project began, I have discovered how information propagates between
insiders, how pricing data on stock is affected by financial news, and how
certain network topologies are useful for representing relationships between
insiders. The most important resources I have discovered are the faculty and
graduate students who have taught me how to form a hypothesis and conduct
research.
