Scenario 1: Yeah, the U.S. dollar is falling, and gold is expected to rise. A look at the gold, sure enough.
Scenario 2: The U.S. dollar has fallen again, and now gold is in the sky. Take a look, hey, why didn't gold go up? It doesn't matter if it doesn't rise, but it still falls with the dollar? ?
I think this kind of situation should be encountered often. That's what I mean by relevance.
What you need to know is that correlation does not mean causation. There may not be an inevitable logical connection between two correlated data. Completely correlated data and product prices are often from the same source.
In other words, if the price movements of two varieties are very similar, it is not necessarily that A has affected B, or B has affected A. They may have a common parent C, and it is C that jointly affects A and B.
Therefore, the conclusion that A is the effect of B or that B is the cause of A cannot be obtained through the correlation between A and B in the data.
The variables of different varieties are different, and the facts reflected by different data are different. There may also be a very close logical relationship between data without correlation. Why? Each economic data, or the data of each variety price, has its own principles and variables. The variables and principles of each data are different, and the facts reflected are also different.
That is to say, it is necessary to understand the fundamental principles of a data, a curve or a variety, and then further obtain its variables on a micro level, and understand what the facts it represents by studying its variables, that is, to understand why the data is rising Up, down, down.
When you can understand a piece of data from the perspective of principles, variables, and facts, you can basically understand it. Therefore, the understanding of data or varieties cannot stay on the correlation, which is very easy to make some mistakes in research and subsequent reasoning and inference.
For example, the prices of asset A and asset B have been similar for many years and many cycles in the past, but suddenly the price of asset B has flattened. Can we draw the conclusion that A will return to B or B will catch up with A?
Let me tell you clearly, no. Because when we cannot understand the principles and variables of A and why it goes up, and the principles and variables of B and why it goes flat, we cannot draw such a conclusion.
Therefore, it is necessary to understand the principles, variables and facts of a variety or a data, so as to compare the relationship between different data and reason whether there will be a regression or catch-up relationship.
The fundamental purpose of studying these is to understand the facts reflected by the data, understand the path of capital, understand the continuous feedback of the behavior of real economy participants and financial market participants, and dig out the logical connection behind the data. Understand the principles, variables and facts of data and varieties through data, and then find their logical connection.
That is to say, if there are similar things or the same parts among their principles and variables, this is the reason for the correlation, instead of seeing that the data are similar on the surface, they are considered to be completely positively or negatively correlated, and cannot Stay on this level.
Simply enumerate several kinds of data, or the relationship between varieties, principles or variables.
The first is the path-conduction relationship . When studying this relationship, its time axis will be moved to the left or right for a certain period, so that we can clearly see how the data is transmitted, because the transmission is time-delayed. For example, A appeared at time 1, and then B was produced at time 2, and B is the result of A. This is a path conduction relationship.
The second is the more common logical causality.
The third is the conditional critical relationship , that is, A and B are necessary conditions for C, and the simultaneous appearance of A and B will trigger C, which is the conditional critical relationship.
The fourth is the mutual feedback relationship . Assuming that the employment data in the United States declines, the government will often increase financial subsidies, which will correspondingly expand the fiscal deficit. The relationship between employment data, government behavior, and fiscal deficit is a mutual feedback relationship.
Looking at the picture below, the purpose of our research is to mine the logical connections behind the data. It cannot be seen that two curves or two economic data are very similar in the front but different in the back, so it is said that this will go down, or that this will go up, and it cannot be concluded in this way.
It is still necessary to understand the principles and variables of each curve, economic data or product price separately, and only then can we understand the economic connotation it represents, infer them separately, and finally draw the conclusion whether it should return.
In many cases, it may not necessarily return, it may continue to rise, or it may go flat or fall.
The purpose of the research is to dig out the logical connection behind the data, including understanding the facts reflected by the data, understanding the flow path of capital, understanding the behavior of participants in the real economy, and understanding the behavior of participants in the financial market.
Only in this way can we find or discover the logical connection in the economy or financial market reflected by different data, including the aforementioned path conduction relationship, logical causality relationship, conditional critical relationship, and mutual feedback relationship.
When a lot of economic data, asset prices, and the behavior of various sectors are continuously further understood at the micro level, the certainty of our future reasoning about prices will increase.
Add
