What we do not do
Recently I have received this question: "Can your software generate buy/sell signals to understand what profit is possible to obtain using your technology?".
This is a very difficult and tricky question. And it is an interesting question. It is interesting to users - as they care about anything that makes their trades profitable. It is interesting for me as well - because there are so many software developers who promise successful trading based on their 80% and more accurate forecast.
I try to be honest. The program is not oriented to obtaining buy/sell signals. It is not because I cannot do that. On the contrary, it is possible, and Timing Solution generates these signals And you can see there a very nice percentage sometimes. However, this issue has many tricky nuances, and the situation is not that simple. Last ten years I have been involved in developing and testing of different mechanical trading systems (based on technical analysis or based on more advanced technologies), so I have had my chances to look at this thing under different angles.
The problem is much more complicated - stock market is a constantly changing system, and we cannot avoid this fact.
Timing Solution allows to generate signals. See this example (here the black curve is the price, and the blue curve is a projection line based on dominant cycles):
The program provides all related statistical information (profit factor, draw down, win/loss, etc). However, I highly recommend not to use this information. At least, you should know how tricky and/dangerous the financial mathematics is. You can get 70 from 100 winning trades (win/loss=70%) within some time period, and the next moment you can lose everything while following the same system. From the point of view of the statistics, your system is very robust trading system (classical statistics tells us that with probability 99.5% this is not an occasional fact), But - financial analysis applies other statistical methods (nonparametric statistics). The gates of Chaos are much more wider in finance, and I do recommend to be extremely careful when somebody promises you a new "robust trading system".
This is a cold wind of reality, and I prefer to be honest with my users from the very beginning.
For your information:
If you are interested in financial statistics, I recommend two books:
1) a classical book "The (Mis)behavior of Markets" of Benoit Mandelbrot;
2) Peters "Chaos and Order in the Capital Markets: A New View of Cycles, Prices, and Market Volatility" of Edgar E. Peters.
Also I recommend the book "The encyclopedia of technical market indicators" by Robert W. Colby; this guy provides a lot of back testing results and shows the real % of performance of the mechanical trading system.
I believe buy/sell signals are really used when you are developing trading robots with rapid trading and very fast trade execution to minimize the slippage. However this is another story.
What we do
We recommend for the trader projection lines adjusted to any
financial instrument. We constantly work with these models improving them.
Timing Solution it is not a software that has been written once; this is a process. All users have access to "Users Area" where they upgrade the software (for free) . We do that because there are no forever working forecasting models; we should improve them constantly taking into account new reality and new information. We do the review of the most working models on our website and on Timing Solution Yahoo Group.
This is the latest (November, 2008) http://www.timingsolution.com/TS/Articles/new_fast_intro/