Our 1Q24 Fundamentals Forecast Performance

Our machines produced forecasts for key fundamental metrics for nearly 1,400 companies of various sizes last quarter. Similar to previous quarters, almost one-fifth of our forecasts were designated as high-confidence. Within this category, chosen through historical error evaluation, our revenue forecast showed an average absolute percent error of 2.8%, surpassing the analysts’ consensus, which had an error rate of 3.3%. Additionally, our algorithms demonstrated remarkable precision in predicting revenue beats (89%). 

Category	# of Forecasted Tickers1	# of Forecasted Tickers w/ High Confidence2	Absolute Forecast Error for High Confidence Tickers3	Absolute Forecast Error for Consensus4	Forecast Accuracy Edge Delivered5	Revenue Beat Forecast Precision6
Overall	1346	259	2.8%	3.3%	0.5%	89.3%
Large Cap	577	112	2.7%	3.2%	0.5%	87.7%
Mid Cap	478	92	2.7%	3.1%	0.4%	87.5%
Small Cap	285	55	3.5%	4.0%	0.5%	96.0%

Our algorithms generate a wide range of additional forecasts at the individual company level, covering non-GAAP gross income, non-GAAP EBITDA, operating metrics, and product segments. Within OdinUltra, the "Key Drivers" sub-tab provides clarity by enabling investors to explore the crucial metrics underpinning these forecasts. Our sophisticated models seamlessly identify key drivers through an in-depth analysis of ecosystem interconnections that surpass human abilities.

To further expand OdinUltra's offerings, we have incorporated aggregate forecasts for 65 industries within the S&P 500 and all 11 sectors of the S&P 500. These predictions encompass aggregate revenue, non-GAAP gross margin, and non-GAAP EBITDA margin. Notably, our aggregated fundamentals forecasts are significantly more insightful than those at the company level.

OdinUltra employs a scientific approach to fundamental forecasting, delivering precise, comprehensive, systematic, and unbiased results. We utilize over 1,400 high-quality factual data sources that are publicly available, such as SEC filings, government agencies, and industry trade associations. We avoid sample/panel data, which suffer from biases due to limited demographic and geographic coverage. Even then, relying on a single data source is purely one-dimensional and can result in significant errors.

Our algorithms have identified over 400,000 direct and latent relationships among the fundamentals of companies, industries, and economies within the S&P 500 universe alone. Each company has a unique fingerprint defined by its product configuration, geographic reach, and industry-specific metrics. Furthermore, today’s companies are becoming increasingly complex and interdependent due to extensive supply chains, expanding global reach, and growing end-market diversification.

For customized strategy construction using our fundamental forecasts into the earnings, please contact us at strategy@odinultra.ai.