Saturday, April 25, 2009

Visual work of the trader

Most traders spend much time on the establishment and optimization of new, highly efficient trading systems. However, the real work of the results are not so impressive. Rather, in most cases blame the human factor. The author tries to visually assess the quality of successive transactions by trader using graphics of Japanese candles. The possibility of such an assessment can help the trader to adjust its strategy in the market

Mechanism sapiens
The main activity of traders practicing mechanical approach to work (when they are not in the market, of course), is to create their own trading systems. Typically, these systems include a model of market behavior on the basis of technical analysis, installation of protective orders and some money management rules, more precisely, the rules of the increase (decrease) positions, depending on evolving market trends. Ideally, when the trading system was created, tested, optimized, etc., the trader must operate automatically, performing in the market system served the team. In an automated, but not automatic. The small difference in terminology leads to significant changes in the work: the system is another link - the trader. Made them the human factor can greatly affect the efficiency of trade. This is compounded by the fact that usually when the market does not use a trading system. It is also possible to use multiple configuration of a system. And the choice is usually also a heuristic nature.

How to minimize harm and maximize the benefits emanating from the trader in the work of its systems? As always, you can try to apply a universal way - to use feedback as a result of real work. To do this, provide a simple method of visual interpretation of its results.

Samurai method
As this method of interpretation of the work can be scheduled to invite the Japanese candles. Suppose that a trader has a certain history of transactions on a real instrument. Let us try to interpret each complete transaction (buying and selling) as a standard candle of tehanaliza. The analogue values Open will be the value curve arrived at the beginning of the transaction. The analogue values Close - the real value of the curve at the end of the transaction profits. The analogue value of High will be the value of the profit curve, if the position is closed at the highest (for bovine market) point on the interval of the transaction. A analogue Low - the value of the profit curve, if the position is closed at the lowest (for bovine market) point on the interval of the transaction. In the case of short positions need to change the High and Low. The proposed data representation body candles will demonstrate a real profit from the transaction. Lower the shade - this is a possible drop in profit in the most disadvantaged in the closing of the transaction interval, the upper shadow - the amount of lost profits on the deal before the interval. By placing such a candle in a manner consistent with the actual operations, it is possible to obtain a fairly complete picture of the profitability of the curve, to assess the quality of the trader, the trading system and may further adjust the trade on the real market. After analysis of all individuals in the instruments can be easily evaluated as the applicability of the system and their own mental preferences.

How it looks
Consider as an example of testing one of my trading system on a real trade (Table 1).

Originally intended to test the work of sound, without a stop signal, without the margin, without correction of fundamental analysis. To simplify the results excluding brokerage commissions, overhead expenses and the discount for the term of the transactions. The purpose of operations, it was not making a profit as such, but development of the interaction between the trader and the actual market. It was interesting, because pure system shows the best, both qualitative and quantitative results than with the limitations imposed on it by the foot and the various filters.

Apart from dates, at first glance, everything is fine: the profit curve is going up. Interest earnings on the use of capital is also not bad. Let's see now, as the profit curve is a graphical interpretation (Fig. 1).

There are already the outcome of the case. Transactions 1-5 were quite successful, but I decided to play with time window and attempt to increase profits. The result in the absence of a stop-signal, demonstrating a significant Drawdown bar number 6. Bar shows the number 7 is nothing but sheer stupidly attempt to go against the market, despite the mounting losses, ie averaged.

Interestingly, the trading system has proved quite resistant to the most unprofessional actions of the trader and drew the final profit. Again: the transactions were carried out for debugging trading system and it was assumed that the human factor will not be. But, alas, even in small-money, in the experiment, all turned out differently.

During the experiment, I broke a lot of work in the market. Bar 3 - spasmodic closure of the position without the alarm system, 2 bars - the closure of positions in the movement are not, the bar 6 - to change the parameters of the system directly to the work of child illness averaging mentioned. Really good, you can recognize only the bars 1 and 5. A further development of the proposed method may be operative in the course of bidding, the analysis of transactions and the selection of appropriate strategies for each trader. For me the best are the candles with small upper shadows. However, this topic has already applies to the consideration of characteristics of my trading system and not connected with the proposed methodology.



Amir Tayupov

No comments: