Detecting a dependent process by Clifford J. Sherry Ph.D.
If you know the price of a commodity today, can you predict what the price will be tomorrow? Can you predict whether the price will increase or decrease? Dr. Sherry proposes a statistical method that allows you to determine which prices are random and/or independent.
If you know the price of a commodity today, can you predict what the price will be tomorrow? A week
from tomorrow? A month? Can you predict whether the price will increase or decrease? By how much?
The answer to these questions depends on who you ask. Some people, like technical analysts, believe that
you can. Others do not.
One reason why this question is so difficult to answer is that there is no general agreement about how
commodity prices are generated. If the underlying process is random and/or independent, then you
probably cannot 'predict' price movement. But, if this is not true, if it is either non-random and/or
dependent, then some 'pattern' exists in commodity prices. With the appropriate tools, you should be able
to detect the pattern and use this information in your decisions about the market. It is important to realize
that randomness and independence are separate and distinct characteristics. For example, it is possible for
a process to be random and dependent, non-random and independent, etc.