forex trading solutions

TradingSolutions

Revolutionize your trading strategy with TradingSolutions by combining traditional technical analysis with state-of-the-art artificial intelligence (AI) technologies. TradingSolutions uses neural networks and genetic algorithms to learn patterns from historical data and optimize system parameters, allowing you to create highly accurate trading systems. This trading software works with stocks, futures, currencies (FOREX) and many other financial instruments. It can also build systems for U.S. and International markets.

Independent Trader Averages Returns of 80% to 120% per Month for 2003
Yuri Shramenko

Independent Trader
Yuri Shramenko has been a full-time trader since 1989, after working as a software developer consultant. With no formal training other than reading books in the late 80s and early 90s, his trading style is a combination of Fibonacci systems and his own personal systems. One of the books he recommends for those wanting to gain a better perspective of market behavior is "The New Fibonacci Trader: Tools and Strategies for Trading Success" by Robert Fischer and Jens Fischer.
Mr. Shramenko started out using TradingSolutions End-Of-Day in early 2002 and upgraded to the Real-Time level when it was released. He uses TradingSolutions to swing-trade the QQQ single-stock futures contract, and day-trade the Dow-Mini and the S&P-Mini futures contracts. His performance numbers are nothing short of spectacular. He has had an average return of 80% to 120% per month for 2003. About 60% of his trades were winners, which may not sound overly impressive, until you realize that his average winner was 4 times larger than his average loser.
How is Mr. Shramenko able to achieve these remarkable returns? His intraday system for a particular futures contract is based on a pair of TradingSolutions’ neural network models – a long-periodicity model (usually 30-minute bars) and a short-periodicity model (either 2 or 5-minute bars based on the average trading range of the 5 minute bars – when the market has a wide range he uses the 5 minute model). The long-periodicity model is used for trend analysis and the short-periodicity model is used for decision making. This is consistent with a basic premise of technical analysis, that of trading in the direction of the next larger timeframe, using the shorter timeframe to specify entry.
Both neural network models are configured to predict the closing price one bar into the future. The parameters of the neural networks are generally left at the defaults provided by TradingSolutions. A sometimes overlooked feature of TradingSolutions is used, that of using other markets or indexes as training inputs. Neural networks excel at finding relationships between correlated non-linear data streams. A recommendation Yuri makes to all TradingSolutions users is to not just be concerned with what technical indicators they use as inputs, but to see if other markets and indexes may indicate coming trends in the market they trade. The inputs for the long-periodicity model include:
Percent change in close
Advancing Issues - Declining Issues (eSignal Symbol: $ISSU)
Advancing Volume - Declining Volume (eSignal Symbol: $VOLD)
Formula: Moving Average (Close,5) - Moving Average (Open,5) The inputs for the short-periodicity model include:
Percent change in Close
Advancing Issues – Declining Issues (eSignal Symbol: $ISSU)
Advancing Volume – Declining Volume (eSignal Symbol: $VOLD) Some key elements of his daytrading strategy can be summarized as follows:
If the price is greater than the predicted close of the long-periodicity model and a chart shows that the prices are in an uptrend, then long signals from the short-periodicity model will be accepted.
If the price is less than the predicted close of the long-periodicity model and a chart shows that the prices are in a downtrend, then short signals from the short-periodicity model will be accepted.
If the predicted close of the short-periodicity model is in the same direction as the long periodicity model, and the actual price goes beyond the predicted price in that direction, then a trade is entered.
A stop loss order is placed as soon as the trade is made.
An inactivity stop is also used. If the trade does not go the right way within a certain time period (approximately15 minutes), exit the trade with an “at the market order
Positions are never held overnight. The typical length of a winning trade is 25 to 40 minutes. But if three consecutive 5-min bars have higher lows (in a long trade) or lower highs (in a short trade) then he uses a trailing stop to capitalize on price extensions, which happen several times a week in the stock indexes.
Before making the trade, chart the ISSU and VOLD to determine when the market is going to go flat or trending. If they are contradictory (one going up and the other down) then the trade is not made.
Retrain the models once per month. This clearly illustrates how neural network models developed with TradingSolutions can be fundamental part of a very profitable trading strategy. We would like to thank Mr. Shramenko for being so open with the details of his system, and we wish him continued success.