Bulltrack Investment Software Message Board

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Strategies & Market Trends : Bulltrack Investment Software

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next

To: JohnQInvestor who wrote (11)	6/5/2000 4:16:00 AM
From: mark_andrews	Read Replies (1) of 14

Hi JohnQ, BullTrack is based on the old adage that the real fundamental valuation of the market is really in the price movement. The most you can hope for is to be holding the strongest stocks. Trying to figure out what's best is very complex and time consuming. BullTrack combines the best of both worlds to create a buy and hold system that will trade only to move into a better position. From 1970 to 1985, investing was much simpler because all you had to do was hold the stock that was going up the most based on performance since you started investing. Today, there is much more volatility in the market so it takes much more to achieve the same thing. If you used the same tactics as PRE-1985, you will be whipsawed and find yourself on the wrong side of the trade most of the time. We use Technical analysis to smooth the volatility and achieve a model that remains very stable and profitable over a long period of time. What do I mean by stability? The model remains at it's maximum performance over a long period of time moving forward. Most programs are very unstable and some (I.E. Omnitrader) will change recommendations during a trade. 2 years of research went into discovering what it takes to create TA driven models that perform well moving forward, not just during the backtest. Most system developers will tell you that the backtest indicates how the system performed during the test period. This is the furthest from the truth and we believe it is fraudulent to make such a statement. Backtesting is a necessary evil but real life testing is the only test that can indicate performance. The real life test optimizes the model every day during a forward test. (Real life test = ReOptimizing walk forward test on Out of Sample data) If the optimized settings do not change during a 1 year real life test, you have a stable model. This means if you traded this model during the forward test (out of sample) and optimized the model today, the trades for the past year in-sample are the same as the recommendations you received during the year you were trading the system. This means the system knew the best setting and is still as perfect today as it was a year ago. This also means the model will remain very profitable moving forward. Stable models are models you can trust because you know they can not get any better. They are absolutely perfect. Another example is with a Neural Network using a verify period. Imagine training a neural network leaving out the last year of data. Then train a second neural network including the last year of data. After both networks are fully trained, add the last year to the first network. Without any additional training, the trades in both networks identical and all other factors including fitness are the same. Even if you train on the last year of data into the first network, the network does not get any better. This is what I have achieved with BullTrack and I am not using a neural network so "training" is much, much faster. Any questions? It does not get any better than this. Regards, Mark

Report TOU Violation

Share This Post

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next