#Polymarket预测市场 After reviewing Kalshi's research report, the data is quite interesting. Over a 25-month observation period, the prediction market's average error in CPI forecasts was 40% lower than Wall Street's, with even more pronounced advantages during volatile periods, sometimes exceeding consensus by 67%.

The key lies in the mechanism of the prediction market—traders' economic incentives drive the full pricing of information. Compared to unidirectional analyst consensus, the "collective intelligence" formed by market trading responds more sensitively to changing environments. This is not to say that Wall Street is unprofessional, but rather that the two sources of signals have their own characteristics: analysts base their predictions on models and historical experience, while prediction markets are driven by real-time capital battles.

The insight for investment research is that, during periods of high uncertainty (such as policy shifts or data-intensive times), trading prices in prediction markets can serve as a supplementary reference. There's no need to choose one over the other; combining these two types of signals can better capture the market's true pricing of macro variables.

Future attention will be on whether Kalshi Research will open up more data dimensions and how this framework performs with other economic indicators.