Trading in the FOREX market is a global phenomenon, and we recently traveled to Santiago, Chile, to talk with Professor Manuel Madrid-Aris on the trading system, which he develops for Forex markets.
Dr. Mardrid is an economist and also has a degree in civil engineering. He has been involved in international consulting on economic development. Over the past few years, he used his experience to develop a trading system for the Forex markets. We asked him several questions about how he uses his academic expertise in trading strategies.
Question: You have developed a system to identify signals in the FOREX market? You can describe what the components of the system?
Manuel: Our system tries to predict price changes in advance. Our system was developed using as inputs the results of the analysis of historical data for the different currencies, using multiple econometric studies. Predictive power of our model is based primarily on correlations of past experience. So, if something happened in the past, it is very likely to happen in the future, especially those effects that are based on structural issues, and price deregulation.
Our strategy is a strategy with the two approaches, a mixture of elements identified econometric analysis, supplemented by technical analysis indicators. Given the fact that currency markets are not perfect and efficient markets, the historical time-series analysis gives very good hints about the behavior of some currencies in different circumstances. In our model, the results of econometric studies, as well as elements of technical analysis used baseline data to make investment decisions.
Econometric studies identifies various facts such as (a) structural changes, based on intra-day, day, weekly and monthly analysis; (b) the variability in different times of the day, based on demand; (c) historical patterns of seasonality of various currencies; ( d) the dynamics of integration among various currencies; (e) external influences (demand or supply) and the elements of timing, which create price deregulation. On the other hand, methods of technical analysis, trying to filter out market noise from the time series, determine the momentum in the market, identify trends and Facing define some standard key patterns, among other facts. Thus, our program relies on a multi-system methodology, which combines several market factors, to identify the most promising opportunities for the opening of trading positions. Definitely, our model tries to take advantage of short-term deregulation of prices.
Through our fully computerized system and the application of our risk management policy, in each transaction, we are trying to minimize losses. Our system is a dynamic trading system that combines, among other elements of the historical behavior of various currencies, which is a relatively new concept in trading systems.
We are convinced that this is one of the most advanced trading systems for the Forex market. Our system was not designed to sell trading signals. It was developed to manage our trading account.
Question: You are using advanced mathematical, fractal, or a more traditional analysis?
Manuel: We used historical econometric analysis to determine the historical common model (eg, structural breaks and timing) for the different rates and ratios of rates (the dynamic correlation among currencies). Thus, we use math in the process, applying the econometric study.
However, these results are included in our package in a simplified manner, without complex mathematics. Our program works, mainly by using simple equations and specified patterns. Specified patterns schemes like «if - that» very common in our system. For example, if X and day X time of day these elements are met, then check other facts, if all the facts are correct, then commit the transaction.
Question: How do professor of economics, you are using economic knowledge in the system?
Manuel: The use of economic knowledge in the trading system can be divided into two types of economic knowledge: (a) macroeconomic variables; (b) micro-economic behavior (the expectations of people.)
Macroeconomic analysis, which largely examines the various macroeconomic variables that affect the Forex markets, such as current accounts, merchant accounts, interest rates, etc., are not included in our system. This is a known fact that there was a time lag between macroeconomic policies, which has been implemented and the time needed to it has had any effect, or the publication of new data as it does not affect the price. This time lag may take a relatively long period of time. In addition, the relationship of macroeconomic variables in different currencies are too complex. The inclusion of macroeconomic knowledge makes sense for long term trading systems, but our system is very short-term trading system, and the inclusion of macroeconomic variables (or known as fundamental analysis) does not make sense. Thus, macroeconomic variables are not included in our system.
For our short-term trading systems, an average of 85% of our transactions - this is intra-day transactions, 14% of the positions held by one to two days, and only 1% of our orders remained in the game longer than two days. None of our positions do not remain on the market more than five days.
On the other hand, the economic analysis included in our model as a micro concept perekuplennosti and pereprodannosti and the establishment of price deregulation in the short term. The key point of our model is the use of the above concepts, supplemented by the core technical analysis and calculation of the seasonal variables that determine the entry point to minimize the risk.
Q: What features of the system for managing risk?
Manuel: Actually, the way that our system is structured, allows us to deal with very close stop-order and very quickly the system determines whether the trade in the direction of where we expected. We are united in our system, several key elements in the risk management module to reduce the chance of Losing Money. Our risk management involves many aspects.
With regard to the major markets in which we sell, we sell only the major currency pairs like EUR / USD, GBP / USD, USD / JPY, USD / CHF, USD / CAN, and GBP / EUR. Thus, we are dealing only with currency pairs have a small spread and high liquidity.
As for the maximum number of simultaneous transactions. We have simultaneously a maximum of three currency pairs.
We use a little leverage to reduce risk. Leverage is from 1:2 to 1:10. Leverage pre-determined before the transaction, based on the ratio of return to risk.
At the level of risk, we will never risk more than 2.3% of total assets.
Regarding the ratio of return to risk. A key parameter to be successful in trading on the FOREX market is the ratio of return to risk. If you do not get much profit, what you risk, it is only a matter of time when your assets will end. Typically, many traders recommend that the ratio of revenue to the risk was high, but there is an optimum point between the ratio of return to risk and the number of transactions from the winning total transactions. Some traders have a higher ratio of return to risk, but low percentage of winning deals, therefore, it can lead to what they have become unprofitable.
On the ratio of the winning and losing deals. Our win-win transactions constitute 66%, losing the deal - 24% and zero transaction - 10%. Many zero-transaction is the result of movement of our stop-orders after the market moved to break-even point.
In the case of stop-orders. Stop orders are always placed after the opening position and remained until its closure. Our stop-order and determined the average daily volatility and variability in the time of the warrant and the average target for the transaction, as defined previously. For short-term trading decisions (intra-day transactions), the average stop-order is changed from 13 to 26 points. For long-term positions (one or two days) a warrant varies from 35 to 50 points. The freeze order will not change much in the way of the original level, even if the market moves against the projected price. Thus, the initial pre-defined level of profitability to the risk of never getting worse.
Regarding the temporary format. To reduce the risk, we use short-term trading. Most of the positions is the intra-day. Thus, the market and reduced risk of accidental, as the majority of positions entered and closed within one and the same day.
On the modeling and the amount of positions. We conduct periodic historical simulation to determine the optimal size of the positions.
Question: How does the system take account of news events? Not whether they will violate the security of algorithms?
Manuel: Our system is dynamic, and the advantage of the news event. Usually, after a news event (such as the war in Iraq), there is some pereprodannost perekuplennost currencies and other currencies. At this point, price deregulation will be present in multiple currencies, and then our system uses these effects to their advantage. If at any time, any news event occurs, which violates the direction of our current position, the position will be closed, and the expected loss will be minimal, by establishing a close stop-order. Then the system can identify the entrance to a new direction.
Question: With whom do you work in a team to develop a system?
Manuel: Designing the system was directed by two economists, myself and Rodrigo Garcia. During the development of the system, the team worked the other economists, as well as two programmers. In addition, we received very helpful advice from the - from Frank Konlina, who explained to us the practical, operational and structural aspects of Forex market.
Question: What are the measurements you use to assess how the system works?
Manuel: The system was tested using data of past four years. Testing was conducted using a small leverage (only 1:5). According to the results of testing, the average annual yield is 88%. According to the maximum monthly loss in four years amounted to only 3%. On average, 90% of months were profitable. Test results show that the inclusion of strict risk management reduces the likelihood of large losses. Next month, we will begin to trade «live», and our system will be evaluated on the basis of daily and weekly periods. The criteria we use to evaluate our system as follows: the ratio of the winning and losing transactions, the average yield, the average number of points per winning transaction, the average number of lost items to a losing deal, and the ratio of return to risk. These parameters will be evaluated for each currency pair. Thus, we can adapt our trading strategy in accordance with individual returns, which gives each currency pair to maximize the total profit.
No comments:
Post a Comment