A London-based experimental prediction market sees the weak El Nino conditions in July falling to almost neutral levels in August before picking up gradually in the next two months.

The Hivemind El Nino-Southern Oscillation (ENSO) Prediction Market made these projections based on the predicted value of sea-surface temperatures (SSTs) in the Central and Equatorial Pacific during these months.

The predicted value of SST anomalies for July is 0.80 deg C, with a standard deviation of +/- 0.57 deg C, which is slightly above the threshold for declaring El Nino conditions in the Pacific.

EXPERIMENTAL MARKET

The Hivemind ENSO Prediction Market is an experimental prediction market to demonstrate how prediction markets can be used to elicit and aggregate the judgment of experts into a probabilistic consensus forecast.

Markets for SST anomalies in the Central and Equatorial Pacific from July to October 2019 are currently open and further markets for November 2019 to March 2020 will open next week.

Mark Roulston, co-founder, Data Science, Hivemind, told BusinessLine that experts in seasonal climate prediction are participating by trading contracts on Hivemind's Agora platform that covers a specified range of SST anomalies in a target month.

Prediction markets can be applied to climate and environmental variables, allowing organisations to obtain probabilistic predictions of variables to which they are sensitive.

The table shows plots of prices for the individual months. The prices can be interpreted as the probabilities that SST anomalies will fall in a particular interval (the intervals are 0.1-deg C wide).

FUTURE UNCERTAINTY

The expected values in the table are the mean values of these distributions and the standard deviations are a measure of their width, so they effectively provide a measure of the uncertainty of the forecast.

The months further in the future have higher standard deviations because there is greater uncertainty.

The final two columns of the table give the probability that SST anomalies will be above +0.8 deg C or below -0.8 deg C, corresponding to El Nino and La Nina conditions respectively, said Roulston.

“So, at the moment, the market is implying a 0.46 probability that El Nino conditions will prevail in October and only a 0.06 chance of La Nina conditions.”

The 0.8 deg threshold for SST anomalies is broadly in line with convention, says Roulston. As for the last two columns in the table, the probability that an event will occur is expressed as a number between 0 and 1.

Probability distribution of monthly SSTs in Equatorial Pacific implied by current prices