All traders do it

How to increase the effectiveness of the various technologies used trader practitioner? Vice President of Business Development <Franklin & Grant> will start a series of articles designed to help make sense of this delicate matter.

Control theory

What strategy speculation take, all traders, somehow, solve a very similar problem. If we try to classify them, it becomes clear that the main types of problems is not so much - only three.

1. Stock selection that best fit the strategy of your game.
2. Forecast future price movements in the interest period.
3. Exhibiting specific orders.

These tasks correspond to the three main stages that are well known in the theory of management: the collection and analysis of information, the forecast of the situation, decision-making for the correction in case of deviation from the predicted dynamics. At this similarity could not pay attention to manufacturers of analytical software. Thus, to select the purchased shares gained wide popularity Stock screener, or a simple filter - from the database they choose only those who share your settings. Price forecast for the widest range of instruments is used. These include traditional methods of extrapolation, and rather complex (and expensive!) Solutions based on neural network algorithms.

For issuing orders to the market, there are many trading systems. Most of them allow you to program the decision rules for the automatic placement of orders. But the rules themselves should ask user! Work to make such rules automatically take over the mechanical trading systems, but most of them prove to be unfounded.

What emerges is a picture? Software products and online services are based on the theory of management, but of the impressive effects that accompany its use in traditional areas for investment management is clearly not observed. What's the matter?

Simplicity is worse than theft

A deeper analysis of the proposed tool shows that in most cases, the program is based on analytical unduly simple algorithm. This simplicity does not match the sophistication of the object - the market. Judge for yourself. For example, what are the stock-screener? This is just one query to the database with the correct <profile>. For market professionals who know exactly what they want, such a tool would certainly be useful.

But it is much more relevant, in my view, would be a tool to determine the permissible value ranges for each metric that characterizes share, for specific purposes. For example, the user indicates that he is interested in weekly growth companies. In what range of interest should be based on the values of its parameters, the probability of this growth was the highest? Now in the market there is no tool to answer this question.

Or this aspect. To specify the stock-screener, you need to know what the value currently on a specific group of companies, for example, drug companies generally are. What many companies falls into a certain range of values of a particular indicator? Unfortunately, this issue is resolved satisfactorily. Recently, attempts to build such a <map> market, but not in all the sources, and only on one or two indicators, such as profitability and capitalization. What to do when you want to monitor multiple indicators?