Trading strategy
A trading strategy (also ambiguous: trading system ) is a set of rules for the trading of securities or futures. A distinction manual and automatic execution of strategies. The latter are referred to as mechanical trading systems.
Manual trading systems consist of a few simple conditions and instructions that can be performed "by hand". Mechanical trading systems can be very complex algorithms included and will be executed by a computer.
Most commercial systems rely on either fundamental analysis or technical analysis and indicators to generate entry and exit signals.
Typical trading systems technical analysis
The typical technical trading systems can be divided into the following classes:
Trend followers
Trend-following trading approaches try to enter existing price trends. They get out as soon as the trend " breaks ". Because it is naturally impossible to identify a trend before it has formed, trend followers often also called "free riders ". You take it in stride, not to go along the entire movement, but only part of it. Global sequence has nothing to do with techniques based on the anticipation sought trends.
Trend-following systems have been known by successful traders like Richard Dennis and William Eckhardt in the managed futures scene. Due to the spectacular story of an experiment in the early 1980s, the Turtle Trader System gained worldwide notoriety. It was first introduced in 1993 fully disclosed and published.
Pullback
A pullback trading system waits for a counter- movement in an existing trend and then increases in a trend direction.
Channel Breakout
It defines a trend channel. Leave the courses the channel, the system increases accordingly.
Cycles
This approach assumes that the price movement cycles are included. There are, for example, seasonal fluctuations in the prices of raw materials. Well-known examples are the 6-phase model of Leon Levey or the " Egg of Kostolany "
Pattern
When trading patterns (called patterns), it is assumed that there are certain, also in the future repeating patterns in price of a security is, as market participants act involved in the same situations the same - so the assumption. Examples of classic patterns are triangular formations, banners, rectangles, double top and double bottom.
Data base
For the operation and testing of trading systems, the historical price data, possibly even volume data and business-related messages of a security are needed. A distinction is made between different time frames: " End of Day" ( EOD ) data take a trading together in a record. The so-called " intraday " data, however, have a resolution of hours, minutes or even ticks.
Strategies of Fundamental Analysis
- Value Investing Stock Picking
Strategies in high-frequency trading
For high-frequency trading, there are specialized strategies that exploit certain effects that occur on this very short time level.
Development, backtesting and real simulation
Development
Electronic trading systems are implemented in software, therefore there exist various specialized platforms and programming languages.
Backtest
In the so-called backtesting the system is tested with data from the past and possibly the parameters with respect to certain criteria, such as Performance or volatility optimized. In the optimization, the problem arises that is unclear to what extent the trading system is actually improved or whether the trading system is simply better adapted to the available historical data. In extreme cases, be adapted to the price history algorithm is indeed deliver good results in backtest, but on new data no longer work.
The validity of a pure " backtesting " is thus low.
Forward test
By optimizing on historical rates still no indication of the profitability in the future can meet. That's why trading systems must be tested after completion of the development and optimization of new and yet unknown price series. This is called accordingly Forward testing.
This test represents a fundamental feasibility test, and will examine a trading model in terms of the assumptions made.