One web site of choice is "Bank Underground" from the Bank of England. Essentially it is about how it works and tries to explain what it is up to or not up to as the case may be. It is neither fun nor easy.
Certainly, it needs a site like this because it is all too evident that most or nearly all of the main media, political parties, traders, dealers, retail bankers, experts of one sort or another and far too many economists are not really up there with the economic game.
To be fair, the game is not the old fashioned single entity where the rules are more or less the same from year to year, and there is a fair chance that predictions may be right or work. In effect, the rules change almost by the day as well as the pitch, the players and the purpose.
This article in titled "New Machines For The Old Lady" is about the advances made at the Bank of England in applying high and new technology to its function as a central bank. There has been, it says, an explosion in the amount and variety of digitally available data.
All you need are machines that will analyse it and allow you to suggest the policy options it alleges are required. If you are in a hurry with all those berserker politicians crying for answers, it seems a good idea.
I prefer the Bank articles to be brief and not to challenge the wiring between the ears. This one needs time because of the subject matter and having to explain what is what. But if you want to know what your central bank is up to, why and how, it is part of the answer.
Unluckily, in this world however good the mathematics, science, data gathering, artificial intelligence, analytic systems and coffee machines, there are no certainties and not much comfort. After all the explanation, it ends:
However, care is needed when interpreting the outputs from ML models. For example, they do not necessary identify economic causation.
The fact that a correlation between two variables has been observed in the past does not mean it will hold in the future, as we have seen in the case of the artificial neural network when it is faced with a situation not previously seen in the data, resulting in forecasts wide of the mark.
Told you so.
Told you so.