I'm in Los Angeles and wondering if there is anyone interested in collaborating on a new research effort using Matlab to analyze data and build a trading system? I am a Matlab expert and have been using the tool to research price action and develop trading systems. If you're interested, we just might be able to accomplish a lot together. Brad
Hi Brad, My name is Mike and I live in Chicago and would like to talk with you about your post, I think it is a great idea. My e-mail is aov@sbcglobal.net. Hope to speak with you soon.
my question was for the OP, anyways I do non directional trading with options selling and directional with trend following
Brad, Just curious. Matlab is something I've never gotten into but have always had an interest in. My day job has me performing data mining on very large data sets. I assume that your education in Matlab was actually non-trading related correct? If so, how did you get your start in Matlab, and what do you consider expert level? Regards, John
Hey this is great. I didn't know if I would find anyone interested. I'll try to respond to several points in this post. I am a radar data analyst and radar detection algorithm developer. I do research using mathematics to program algorithms for various radar applications. Although the radar work involves somewhat complicated math, my trading is more about technical analysis and KISS (Keep It Simple Stupid). I built my own backtesting platform in Matlab and have used it for my research. I have many areas of interest. Current areas of interest include: Seasonality - from intraday up to the 4 year presidential cycle (BTW, this appears to be a form of trend following if it is implemented successfully) Reversal Pattern (morning/evening star) Trading into the close of the SP500 (also trend following) Trading the open of the SP500 Relative Strength These are the most important to me right now. I have hundreds of ideas documented that include gaps, breadth, volume, Dow Theory, breakout, swing points, etc. I have a very broad trading education, a scientific background, too many ideas, and trouble focusing on one thing. I have proven several strategies true and continue to do research in multiple aspects of trading. I am indifferent about which type of strategy I study. I believe they all work, but a lot of research has to be done to learn how to implement each. Ultimately I want to run multiple strategies on many assets in various time frames. I do have a thorough understanding of derivatives but have decided to stick to only long and short positions for now. Also, historical options price data is harder to get. I am interested in studying the relationship between implied and historical volatility. I think market neutral short options would be a painful research project in Matlab, but trend following is ideal. I was first introduced to Matlab in a college digital signal processing class and now use it exclusively at my job. I just started investigating how to connect Matlab directly to TD Ameritrade's API. Matlab is an extremely powerful tool for research and data analysis. The language is relatively easy to learn and the documentation is excellent. Although I have a command of the programming language, I suppose the expertise is more related to understanding how to approach a scientific problem and conduct unbiased research. Brad Trend following is great, ever try panic following?
No, they don't all work. Very few do. If it were that simple, then the hedge funds, investment banks and others with their billion dollar research budgets would be well ahead of you and your Matlab experience.