Artificial Intelligence Is Beating the Professional Stock Pickers

In the 2022 Makridakis forecasting model competition, the winner is AI.

The Makridakis Competitions (also known as the M Competitions or M-Competitions) are a series of open competitions to evaluate and compare the accuracy of different time series forecasting methods. They are organized by teams led by forecasting researcher Spyros Makridakis and were first held in 1982.

Nassim Nicholas Taleb, in his book The Black Swan, references the Makridakis Competitions as follows: “The most interesting test of how academic methods fare in the real world was provided by Spyros Makridakis, who spent part of his career managing competitions between forecasters who practice a “scientific method” called econometrics—an approach that combines economic theory with statistical measurements. Simply put, he made people forecast in real life and then he judged their accuracy. 

Makridakis and Hibon reached the sad conclusion that “statistically sophisticated and complex methods do not necessarily provide more accurate forecasts than simpler ones.”

AI Is Now beating the Forecasters

Eurointelligence notes AI Is Now Beating the Forecasters, emphasis mine.

We have been writing a lot about forecasting failures. We came across a wonderful recent story that should have deserved more attention. Admitted, it did in the narrow world of machine learning and artificial intelligence. It is about the Makridakis forecasting model competition, named after a professor from Cyprus, the latest version of which focused on the stock market. The headline news is that a new generation of modern forecasting tools has gradually been emerging as the most successful over the years. Of the 163 entries, the winner is a model based on a very new method, called meta-learning, which one can translate as learning how to learn. They don’t have a method, they pick a method. The second-placed entry follows last year’s trend that showed the power of another approach, which goes by the name of gradient boosting.

We would like the organisers to go for an economic forecasting competition next time, though we doubt that the central banks and other official forecasters would willingly participate in a competition that risks highlighting their failures. Our hypothesis is that modern machine learning methods would outperform them, and do so by a wide margin. This is quite astonishing because there is absolutely no economics input in those new forecasting methods. The winner of the stock market competition was totally ignorant of financial economics.

The main reason for the relative success of the new generation of forecasting models is lack of bias. Forecasting is very difficult. It is hard to beat the dart-throwing monkey dumb forecast. In the machine learning competition, the dumb forecast occupied 12th place out of 163. Another old-school model, that simply used implied prices from the options market, came eighth. You can think of this as the efficient-market-hypothesis entry. That raises the question of whether the top seven are merely there because of luck. After all, more than 150 entrants were not so lucky.

If forecasting is your thing, then you are about to get a lot of competition.

Another Heavily Revised Housing Starts Joke of a Report

Forecasting is all the more difficult, when there are constant wild swings and revisions to the data. We see that every month in the housing starts and new home sales estimates.

I wrote about that earlier today, referencing a post from last month that reported the same thing.

For discussion, please see Another Heavily Revised Housing Starts Joke of a Report for May 2023

R* the Neutral Interest Rate

John Williams is New York Fed president.

Regarding terrible Fed models, and complex ones to boot, please see Interesting Groupthink Dovish Nonsense from the New York Fed on Neutral Interest Rates regarding R*.

R* Is Not Back!

R* isn’t back, because it never went away. 

There is indeed a neutral rate, but the Fed has done a pathetic job finding it. 

Greg IP is certainly correct “John Williams’ conclusion that the neutral rate is still low raises more questions than answers.”

Fed Groupthink

The Fed will continue to miss finding R* because they have no idea the net impact of de-globalization, decarbonization, and demographics. 

The Fed claims to be data dependent but they stick to group-think models that do not work.

Regarding groupthink on inflation expectations and the Phillips Curve, please see Inflation Expectations are Crashing. So What? It Doesn’t Matter.

Groupthink is alive and well, complete with nonsensical models. It’s no wonder the Fed is so terrible at forecasting inflation.