What are the business drivers behind greater adoption of representation learning in financial markets ?
We have terabytes of data today in machine readable form.
These are all organized in very small packet sizes, leading to billions of discrete events that a trading system needs to process.
With thousands of stocks, ETFs and derivatives, it might feel like the inherent complexity of predicting financial markets is just too high.
The dominant method of trading is to generate features that capture some genuine market intuition and then test hypotheses of how to use them. Feature engineering needs a great representation of data to work well.
Performance of machine learning algorithms depends on the representation of data that is given to them.
In fact, that what we describe as intelligence is really about coming up with a better representation of data. Making inferences from it is often much easier. Those with better intuition just "think differently" about the problem.
Ankit, who leads a global trading operation at DV Capital LLC, will talk about manifold learning. Manifold Learning is a method of figuring out really how markets move. They really move in very few ways and the best traders have learned to see markets in this manner. We can use manifold learning to do the same.
Let's learn together
Our aim here at FinTech - DataScience - NY/JC meetups is for people to come together and share with others what they have learned, towards building a better future.
If you want to want to present at our meetups please email at community@qplum.co