The formation of systematic trading strategy can usually be carried out in the following two distinct ways: top-down and bottom-up.

Top-down refers to forming a certain theoretical understanding through long-term observation of the market, and then forming a certain strategy and tactics based on this understanding.

For example, Dow theory is the earliest and most complete theoretical summary of the characteristics of stock price fluctuation in the early 20th century. Dow theory can be composed of twelve basic definitions. Among them, Dow's definition of trend is the most appropriate. Later, a series of famous trading ideas were formed based on Dow theory, such as "swing trading method" and "three-point trading method" and so on.

The so-called bottom-up refers to finding the corresponding trading strategy from the statistical data of the market and the statistical characteristics obtained through statistical analysis.

For example, "buy after a gap opens a few points, and sell at the closing price" is the so-called "gap effect" found according to the research of American futures database at that time, that is, the probability of price increase after a gap opens a few points. So far, the theoretical circle has not reached a conclusion or a consistent explanation about the reasons for this market phenomenon. This is a typical example of trading thinking formed purely by data.

Whether it is a top-down trading strategy or a bottom-up trading strategy, there are famous successful examples and successful investors in history. For example, baruch, B.Graham, W.Buffett and J.Templeton, famous American investors, are all famous investment theorists of the same era. They all have profound attainments in investment theory, and some even made outstanding contributions to the development of investment theory.

But from the perspective of systematic trading, the idea of forming a trading strategy from top to bottom has the following advantages compared with the idea of forming a trading strategy from bottom to top:

Original: Web page link

First of all, we can well grasp the relationship between local losses and overall failure.

Systematic trading emphasizes long-term stable gains, not gains and losses in one place at a time. Because the distribution of statistical samples is difficult to balance, sometimes the occurrence of adverse events is clustered, that is, any trading system must be prepared for continuous failures. In this adversity, in the face of strong psychological pressure, if traders fully understand the investment theory on which the system is based, it is very possible to clearly evaluate the statistical characteristics of each sample, so as to judge whether this is an adversity period formed by uneven distribution of statistical samples or a new fundamental conversion ratio has occurred in the investment market, so that the trading system can adapt to the new market characteristics.

From many years of trading practice, the author realizes that it is of great significance for traders to keep psychological balance in adversity to fully understand the investment concept on which the trading system is based. Some investors spend a lot of money to buy the trading system for commercial sale, but it is easy to give up the trading system in the face of continuous adversity, and soon after that, these investors rediscovered the system and resumed good performance. There are usually two problems here:

1. Any trading system must embed the psychological characteristics of the system designer. If the psychological characteristics of system users are very different from those of system designers, then the trading system is not suitable for the users.

2. For the problems mentioned above, because commercial trading systems often don't disclose their design ideas and important parameters, users can't fully understand the system from the theoretical level, which eventually leads to psychological imbalance in adversity.

Second, it can be beneficial to the risk control of the trading system.

If the trading system does not have programmed risk control rules, understanding the investment concept of the trading system can make users have a certain foresight on the degree, scope and time of risk occurrence, thus enhancing the initiative and accuracy of risk control. For example, if the user uses a trading system with the design concept of trend tracking, then the user can know that the adversity period of the system is a period of sideways fluctuation through theoretical research and long-term practice of trend and non-trend markets. If the user judges that the market will enter a shock period according to a certain method, then the user can control the risk by reducing the amount of capital investment and observe the performance of the trading system.

If the transaction result does increase the number of losses, it proves that the judgment is correct and the protection measures are appropriate; However, if there is no obvious change in the number of losses in the trading results, it proves that the judgment of the market situation is wrong, and users can gradually increase the amount of capital investment.

Third, it is more conducive to the maintenance and modification of the trading system.

There is a contradiction and unity in the use of the trading system, which is quite difficult to handle, that is, we must unswervingly hold all the signals of the trading system in a certain period of time, but also be alert to the deviation of the trading system from market characteristics, so as to modify the trading system when necessary.