There may be several situations in selecting which causal variable

1，We can construct a variable that mirrors itself. When it sees its own state at the last moment, it determines whether the current state is T or F

2，We can also construct a variable that is numerical, only related to its last moment, so it may become divergent and infinite. Or it may be locked in a value.

We need to consider avoiding those values in the model

3，We can also construct variables that are related to the original time and have a positive weighting relationship with the QM probability of the previous time,

In the case 3 of only one variable, QM probability wrench may not be negative, 0 < r < 1,

This scale of growth sometimes collapses, but it still grows in the long run.

But it may not be true.

The real situation may be like the stock market index. People will depend on the last moment of joy or panic,And expectations for fantasy

until the big external forces play an end role.

This may be a good analysis object

In these analyses, we can divide the current value by the above one to obtain a random numerical sequence

**Cellular automata** has demonstrated that in an incomplete observation process (discrete time and space),

Random looking sequences can also be constructed from regular transformation directions

In a complete system process, once all paths in the space are traversed completely,

These sequences will restart the previous path

Replay everything. (I've experimented with this in one complete cellular automata.)

I think the probability in QM may also have this rule.

In other words, probability itself exists in causality.

When we do not consider the above, we can also regard QM wrench as a variable.

If so, it will continuously change the tensor matrix of time and space, and pass on the changes,

From a global perspective, this is no different.