Capital management strategy in a mechanical trading system in the market
Kryukov P.A., Kryukova V.V.
г. Kemerovo
The dynamic indicator under consideration is loss coverage ratio as an element of the structure of the mechanical trading system, which allows you to trade with a minimum loss by implementing situational control over the management of currency positions.
Introduction
In the case of a mechanical trading system (MTS) strategy rules are formulated (formalized) with the participation of mathematical models (formulas), it is very important to determine the moment when there is a need to update the parameters (coefficients) of the model equation, to save own funds in the process of trading as much as possible, as well as to exercise situational (operational) control over the trading results. For this purpose it is suggested to use dynamic indicator - coefficient ktwhich reflects the current efficiency of trade and shows the ratio of the excess (shortfall) of cash inflows over outflows (personal contribution of the author Kryukov P.A.) in the process of trade.
Competent money management (money management strategy) in the developed MTS [1] involves forced loss limitation, namely, the opening of a position in the presence of a given share of the initial deposit in the account and the immediate closing of unprofitable positions on the combination of signals model (the basis for formalizing a trading strategy) and the dynamic indicator - loss coverage ratio.
Capital management strategy
The basis of the strategy in the developed MTS [1] is a dynamic indicator - loss coverage ratio, the structure of which is as follows:
Where the numerator:
NP - accumulated fixed profit by the time t on all closed positions;
N/P - undistributed profit with "+" sign for the current open position;
no - the share of own funds in the account, which the trader agrees to lose (closing risk);
denominator:
NY - cumulative fixed loss on all closed positions at time t;
N/P - undistributed profit with "-" sign for the current open position.
The result shows the current 1% loss coverage.
Rules of the strategy: if kt>1 for the current value of the exchange rate, the position is maintained; if kt≤ 1 OR there is a signal of the model to close, the position is closed immediately. The position is opened by the signal of the model И the truth of the condition:
where Akk - own funds on the account at the current time t;
na is the share of cash from Deposits (risk of opening a position);
Depo - the initial value of the deposit.
Empirical testing of the capital management strategy
To test it, we use the trading strategy formalized on the basis of the model of typological classification of the current trend by the factor scaling method for the currency pair eur/usd [2].
Experimentally determine the values of the indicators no и at for profitable and unprofitable trading variants. At first we will trade on seven price history intervals with optimal parameters of the optimization variant 2328 [2], which are profitable strategies. We examine the following intervals: 1 - 1.06.2004 - 1.06.2005; 2 - 1.06.2005 - 1.06.2006; 3 - 1.06.2006 - 1.06.2007; 4 - 1.06.2007 - 1.06.2008; 5 - 1.06.2008 - 1.06.2009; 6 - 1.06.2009 - 1.06.2010; 7 - 1.06.2010 - 1.07.2011. Let us carry out researches for FSH Strategy [2]. The results are presented in Table 1.
Table 1. Trade results Fsh strategies
Optimal parameters of variant 2328 give strategy profit for different limits of research intervals (table 1; [1], p. 27), which once again confirms stability and suitability of the current trend prediction model developed on the basis of factor data scaling method [2, p. 125].
Adding restrictions on position opening and loss coverage ratio when closing a position, optimization of parameters at и no allowed to improve profitability indicators for intervals 5 and 6: the ratio of total profit/total loss changed qualitatively: net profit for interval 5 increased by 242.9% and by 58.5% for interval 6. For the other intervals the indicators remained the same (Table 2), i.e. adding the loss coverage ratio did not worsen the result.
Table 2. Trade results
The purpose of using the coefficient kt - decrease of real losses in the process of trading. Let's consider the loss-making trading variant with and without restrictions on the interval 27.03.2009 - 1.06.2010 with the following model parameters: p1=0,06; p2=0,12; si1=15,0; si2=14,2 (table 3).
Table 3: Results of unprofitable trading
Attempting to optimize parameters no и atThe profit-generating strategies did not lead to success. However, the inclusion of the capital management strategy based on the loss coverage ratio in the MTS (Table 3) allowed us to increase the "profitability" indicator by 32%, reducing the number of transactions, including loss-making ones.
Using the method of simple search of values, we perform a rough optimization by nodal points in order to find the variant that provides the minimum loss (Table 4, at the intersection of the row and the column is the loss value achieved at the specified values of the parameters). The result for the optimal variant is given in the second line of Table 3. Applying the loss coverage coefficient kt with the parameters no=0,2 и at= 1.0 (Fig. 1) allowed us to reduce the loss by 35.5 times by reducing the number of losing trades (the loss grows - trade stops).
Table 4: Dependence of loss on no и at
Let's study the dependence of the loss value on changes in the parameters no и at. The dependence graph is shown in Fig. 1. The analysis of Fig. 1 allows us to draw the following conclusions.
Figure 1: Dependence of loss amount on at и no
Conclusions
1. When reducing the parameter at (risk at the opening of a position), which corresponds to a decrease in the equity of the account, the loss increases for any values no (risk at closing), and when it decreases - it decreases and the smallest value of loss is achieved if the equity (deposit plus accumulated net profit) - not less than the value of the initial deposit.
2. If the closing risk increases from 40% to 100%, the amount of loss is constant (slightly more than the optimal - -$306.4) if the equity is not less than the value of the initial deposit (at ≥ 1,0).
3. For the indicator risks of discovery items at (equity), changing from 0.1 to 0.9, the amount of loss is about the same (increases slightly) for all values of the indicator no (risk at closing).
4. At the closing risk in 20% (index no) we have the smallest value of loss (-$275,6) if the equity (opening risk) is not less than the value of the initial deposit (at ≥ 1,0).
5. Indicator at has a greater influence on the amount of loss, because it affects the fact of opening a position: if the equity is less than the value of the initial deposit - there is a loss, then trading stops, until all funds are lost. Determining for yourself specific values of risk at opening and closing a position (the values of no и at), the trader can reduce losses by keeping some funds in the account.
The economic interpretation of the loss coverage ratio for the result obtained: when you open a position, keeping your own funds in the account of at least the value of the initial deposit and the risk of opening in 20% of the current value of funds in the account, you can trade with a minimum loss, using the Strategy Fsh.
List of references
1. Kryukov P.A. Development of effective trading strategies in the FoRex currency market // Bulletin of Novosibirsk State University. Novosibirsk State University. Series: Information Technologies. 2011. Vol. 9. Vol. 4. С. 18-29.
2. Kryukov P.A., Kryukova V.V. Forecasting the exchange rate on the basis of factor scaling // Vestnik (Herald) of Kuzbass State Technical University. Kuzbass State Technical University. 2011. №1.С. 118-127.