Leveraging daily survey data from a large pool of respondents and a range of macroeconomic variables, then applying a mix of machine learning methods, occam has shown promise in forecasting macroeconomic releases (10 are being forecasted today, with more on the way). Occam has produced notably accurate estimates for the Index of Consumer Sentiment published by the University of Michigan (UMCSENT). Below, we analyze occam’s performance in predicting UMCSENT and briefly describe occam’s data collection as well as its forecasting approach.
As evidenced by multiple metrics, occam demonstrates a significantly more accurate forecast of UMCSENT than the consensus forecast.
*See data for consensus forecast here.
Source: Analysis based on occam™ proprietary AI-enhanced research platform with various data sources, including a wide range of questions asked to over 1000 respondents per day with over three years of history. Information is census-balanced and uses occam’s™ proprietary AI algorithm that ensures minimal sampling bias (<1%). Contact us for more info.