JPMorgan Economists Tap ChatGPT-Based Model To Detect Trading Signals In 25 Years Of Central Bank Statements

JPMorgan Chase & Co JPM has reportedly unveiled an artificial intelligence-powered model aimed at deciphering the Federal Reserve's messaging and potential trading signals.

What Happened: JPMorgan economists, including Joseph Lupton, employed a ChatGPT-based language model to detect the tenor of policy signals that is based on the central bank statements and officials' speeches going back 25 years.

They effectively rated them on a scale from easy to restrictive and are calling it the Hawk-Dove Score, reported Bloomberg.

"Preliminary applications are encouraging," Lupton and his colleagues wrote in a note.

Also Read: Best Penny Stocks

Signals: The economists found that the AI tool can be useful in potentially predicting changes in policy and signal tradeable signals. For example, they discovered that when the model shows a rise in hawkishness among the central bank speakers between meetings, the next policy statement has become more hawkish and yields on one-year government bonds rose, the report said.

According to the model, a 10-point rise in the Fed Hawk-Dove Score now translates to an increase of about 10 percentage points in the probability that the Fed will hike the policy rate by 25 basis points at the next meet, or vice versa, the report said.

Based on the model, while the reading for Fed statements has fluctuated in recent months and that for speeches has trended lower, both still linger near the highest levels in two decades, a sign that hawkishness continues to persist.

According to the CME FedWatch Tool, there is over 70% probability that the central bank will hike the policy rate by 25 bps in its May meeting.

In the coming months, the Hawk-Dove Score, which is also available for the European Central Bank and the Bank of England, is anticipated to expand to over 30 central banks around the world, according to the report.

Read Next: Cathie Wood Is Unstoppable With Tesla Stock – Defies Wall Street To Make Whopping $29M Purchase