Data Mining on the NIFTY

I'm just back from a business trip in India (Delhi). I went there to meet MarkeTopper, a company that uses data mining for stock market predictions. My first impression was the seriousness of the company. They have an excellent internal structure and their employees are very qualified. Unlike me, they're not making direct predictions on stock market increase or decrease in the future. Indeed, they use data mining algorithms to build their strategies and portfolios.

I can't enter into details about the algorithms they use and how they use it for obvious confidentiality purposes. The meeting was very interesting for several reasons. One of them was the way they approched the problem of using data mining algorithms in the stock market. It is completely different from my personnal approach. I always thought of using these algorithms to predict a value (close price) or a class (increase/decrease) representing price evolution five, ten or twenty days ahead and then applying this process in the past to backtest the system. Their approach is to use the same kind of algorithms to tune a strategy and then backtest it to see its results in the past. Very interesting!

MarkeTopper website

Continue reading... Sphere: Related Content

AI Bot Traders

Botraiders is a french company specialized in individual stock movement prediction. Xavier Bouteiller and Yann Finck use artificial intelligence to predict which stock will increase or decrease in the following days. Of course they do not explain their algorithms, but they are based on data mining techniques among others.

I had the opportunity to meet Xavier in a small meeting in France. His presentation of their system (Botraiders) was very interesting. They have one bot for each stock. Recently, they started using kind of "managing bots" which interact with individual bots to build portfolios.

Up to know their predictions are free and they post them every day. You can have more information about this company and their project on their blog.

Continue reading... Sphere: Related Content

Mining stock market data

After 4 years in the academic world, I will now jump into the industrial world! From beginning of May, I will work in a small financial company in Nyon (Switzerland). I will basically use data mining to help traders invest money. The job seems very exciting, especially since I'm new to the stock market domain. I will of course continue to write posts on Data Mining Research :-)

There will mainly be three kind of changes regarding DMR. First, there will still be posts about data mining, but of course, the applications will more often be in finance. I was applying data mining in civil engineering for 4 years, I will now apply it in finance. However, most of the posts on DMR had a general topic (not related to civil engineering), so it won't change a lot.

Second, posts regarding my current work will be less detailed. When you're in the academic world, you can write in details what you are doing, how you do it and why. The only risk you have (which is ridiculously small to be honest) is that someone steals your ideas and publishes before you. In other words, you can write what you want. In industry, things are different. I will certainly write about what I do, but I will not be able to give precise techniques used or detailed results. It's sad, but that's the rule. And I think this is especially understandable when, like in finance, you deal with huge amounts of money.

Third, the frequency of posts will decrease a bit. I think I will start by posting only once a week and I will see how things evolve. If I can write more, I will do it, of course. I hope that what you will loose in quantity, you will gain it in quality. Finally, in order to start interesting discussions about data mining, your comments are always welcomed!

Continue reading... Sphere: Related Content