Linda’s first research partnership was formed with MRCI and Steve Moore in the early 90’s. Much of this work included 2-period ROC modeling, Taylor modeling, and volatility breakout methods. The work done with MRCI has remained robust, despite numerous changes to the market’s efficiency.
Nigel Bahadur joined LBRGroup in 2001 and was instrumental in refining LBR’s modeling over the next decade. ATR functions are used to model swings and then pattern recognition in the swings became the core of the Fund’s strategies. In addition, original time of day modeling has remained robust, as well as models for persistency of trend, Pinball 1 and Pinball 2, Wide Range reversals, Burning Dog opening gaps, and failed burning dogs. Much time was focused on modeling outlyers as well.
My personal thoughts:
All of our research has remained durable and robust because most of it relies on two parameters and one filter at most. Research is essential but this does not lead to mechanical systems.
After close to 30+ years of research, I have yet to find a mechanical system that I could trade size on with confidence. Either peak to valley drawdowns are too steep, recovery time is too long, or friction (the inability to move size without too much slippage) is too high.
My research and modeling is invaluable to me. I know when things work and when they do not – much is dependent upon placing the markets in a context which is difficult for mechanical systems to do.
If you see tidbits of my research online in assorted places, I hope it inspires you to do your own original work or modify mine. Every trade that I make is based around my models, yet I choose to use discretion on entries. I find systematic management to be where the real gold lies. Human failure due to emotions or sloppiness or neglect, is the real culprit to damage to the bottom line….not lack of viable models!
Think of your own research or modeling as simply a way of asking the market questions.
I would not trust anyone else to answer my questions! Learn to do your own work and you will be a better trader for it. NEVER take anyone else’s research or indicators or statistics for granted. I have seen much flawed work without the author even realizing the inherent problems.
You do not have to have a sophisticated program to do research or testing. For ten years all my work was done by hand. Even printing off a series of charts and marking them up counts as research! Best Wishes!