Hello there,

please refer to this article, I hope it helps

Regards

Article ANALYSIS OF A STANDALONE USAGE AND LIMITATIONS OF A RELATIVE...

According to most academic research no. According to most practitioner research yes. There are technical reasons to not agree with parts of both. I'd stick with knowing smart people who can advise you. At least if you loose money you can blame them. LOL

Prediction of the behavior of a system, at a later point of time depends on how well we understand its present behavior and how this behavior evolves dynamically with time, regression is a type of data driven modeling process where the training data (historical data) is used to build an empirical model for the system behavior in an input-output framework, then validating the model using statistical significance tests and other studies, here an element of learning, i.e. supervised learning is involved, the historical data acts as the teacher and obtained model is the learning outcome which is then tested against a set of test scenarios.

However, this supervised learning process is not restricted only to statistical methods, you can use Fuzzy logic based controllers, based on Mamdani approach or Takagi Sugeno approach and its variants to also to describe the input-output behavior of such systems and the formalism of Artificial Neural Networks to carry out this learning and prediction task, so it is important to realize the more adept the tool used for the learning process the better capability it has in general

In accurately predict the future, neglecting the sample size and over training related issues.

For economic prediction problems, these non statistical tools like fuzzy logic based methods or artificial neural networks, or their hybrid forms can be used as well to see if they fare better w.r.t conventional regression based forecasting methods.

This is a good question. Indeed there is a good scientific basis for thinking about our social system of macroeconomics--although it has not yet been developed to the point of determining prices. I wrote about it in 2015 in only 310 pages of my book "Consequential Macroeconomics". If you write to me at [email protected] I will gladly send for free an e-copy.