How Hypermind predicts the future

The Hypermind prediction market is a forecasting contest on geopolitical, business and economic issues. It is designed to consolidate the informed guesses of thousands of brains worldwide into accurate probabilities. For instance, on the eve of 2020, the market judges that Donald Trump is 51% likely to be reelected president of the USA next November, compared to 40% for another man to replace him, or 9% for a woman. (All our predictions are published in real time here.)

USPREZ2020WHO-EN

As this example shows, the prediction market is not a binary oracle that states with certainty what will or won’t happen. Its pronouncements are more subtle. It doesn’t say that the next president of the USA won’t be a woman, it says that there is currently 91% probability that it will be a man.

One naturally wonders whether these probabilities are reliable or pulled out of thin air. If Trump or another man is elected in november, will it mean that the market was overestimating womankind’s chances to seize the White House? Or if a woman does win, will it mean that the market was severely underestimating her chances? To answer these concerns, one would have to rerun the election a hundred times: if a woman won in about 9 runs out of a hundred, then the 9% probability was correctly estimated. But if a woman won much less often, say only 3 times, or too often, say 18 times, then the 9% estimate was, respectively, 3 times too large, or too small by half. Wrong in any case.

Unfortunately, elections or other historical events cannot be rerun, not even once. Does that mean it is impossible to assess the accuracy of probability forecasts? No, there is another way, less ideal but satisfactory nonetheless: consider allthe events that the market has forecasted with the same probability, say 10%, and count how many have actually happened. If about 10% for them did occur, then the 10% probability forecast was correct. Otherwise it wasn’t. Repeating this procedure over all probability levels, from 1% to 99%, gives a complete picture of the forecasting accuracy of the prediction market.

To perform such analysis, lots of data are required. Hypermind has them. Over 5 years, from the market’s launch in the spring of 2014 until the spring of 2019 (when this analysis was performed), it has generated 537,640 probability forecasts about 1,185 possible answers to 400 forecasting questions in the geopolitical, business, and economic domains. In the graph below, each data point answers the question: « What is the proportion of events forecasted with probability pthat actually occurred? »Amazingly, reality seems to literally align itself to the market’s predictions.

calibration-en

This alignment shows the amazing ability of the market’s collective intelligence to discern, not the future itself, but its underlying probabilities. From this we may draw some conclusions:

  1. The future is not determined by the world’s present state. It is fundamentally probabilistic. It is always wrong to think, after the fact, that whatever happened hadto happen. There always was just a probability that it could happen, and another that it might not. It always could have turned out otherwise if the great dice roll in the sky had landed differently.
  2. Collective intelligence, as expressed in a prediction market, is able to accurately discern the probability field that underlies reality.
  3. The forecasts of the Hypermind prediction market are particularly reliable.

To explore these ideas further, in practice:

Can the Market Set Better Monetary Policy Than the Fed?

FED

For the past year, Hypermind has partnered with the Mercatus Center at George Mason University to try to answer one big question. Here’s how they formulate it on their website:

Currently, US monetary policy is largely based on the Federal Reserve targeting inflation to keep the economy stable. That means it ensures that inflation—the increase in the prices of goods and services—does not venture too far from 2 percent. But for years, inflation has stayed below 2 percent and all the while, we have not seen strong levels of economic growth.

The Mercatus Center’s Scott Sumner and David Beckworth have made the case that an alternative monetary policy approach, nominal gross domestic product (NGDP) level targeting, is superior to inflation targeting. According to Sumner and Beckworth, instead of targeting inflation, the Federal Reserve’s monetary policy should target the rate at which the nation’s total income is expected to grow. NGDP level targeting will ensure that the right amount of money supply is provided to meet the economy’s needs.

But how do you determine if monetary policy is set appropriately to produce stable growth in NGDP? Let’s let the market decide, Sumner and Beckworth hypothesize.

That’s where Hypermind comes in.

We have created a prediction market to forecast NGDP growth over the first two years of Donald Trump’s presidency. It is generously endowed by Mercatus with $70,000 in prizes for those whose predictions are most foresighted. Participation is free of charge and open to everyone interested in the topic, but the prizes are very real! To participate, just sign up at Hypermind.com.

Here’s what the folks at Mercatus are hoping to find out:

The project’s goal is to determine whether the market can adequately determine the trajectory of NGDP growth and whether the Fed should use this information to inform policymaking and to move towards an NGDP level targeting regime. 

In the meantime, click here to view the latest market forecasts (updated in real time).

Political uncertainty, risk of Frexit and European sovereign spreads

politicalRisk

Last year, during the hotly-contested French presidential election, a couple of economists from Banque de France used Hypermind’s trading data to measure the risk of “Frexit” and its impact on financial markets. Their research findings have recently been published in the scholarly journal Applied Economics Letters.

Their conclusion strongly underlines the usefulness of prediction markets such as Hypermind for pricing financially-relevant political risk. It is worth reading carefully:

Two messages emerge from our results:

  • First, predictive markets and crowd-based forecasting appear to produce specific political information about uncertainty and risk that (i) has strong explanatory power and (ii) is not fully aggregated by financial markets, for example through stock volatility or global risk aversion.
  • Second, investors appeared worried by Frexit and reacted strongly when faced with an increase in its likelihood. This suggests that, in accordance to existing literature, uncertainty regarding a specific event weighs on investors’ behaviour, even when its likelihood is low.

Overall, we think these results suggest that the study of political uncertainty spillovers across borders is both an important and promising avenue for further research and that prediction markets will provide useful data for this endeavor.