Our 2Q24 Fundamentals Forecast Performance
By Big Data Federation, Inc.
Last quarter, our models produced forecasts for key fundamental metrics across 1,405 companies of varying sizes.
Similar to prior quarters, approximately one-fifth of our predictions were assigned a high-confidence rating. Among these, identified through historical error analysis, our revenue forecast demonstrated an average absolute percentage error of 2.5%, surpassing the analysts' consensus error rate of 3.3%. Moreover, our machines showcased remarkable accuracy in predicting revenue beats, achieving an 89% success rate.
| 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 | 1405 | 254 | 2.5% | 3.3% | 0.8% | 89.1% |
| Large Cap | 614 | 116 | 2.6% | 2.9% | 0.3% | 88.9% |
| Mid Cap | 496 | 84 | 2.1% | 3.4% | 1.3% | 96.2% |
| Small Cap | 288 | 54 | 2.7% | 3.7% | 1.0% | 78.4% |
Our algorithms generate a wide range of forecasts at the individual company level, covering non-GAAP gross income, non-GAAP EBITDA, operational metrics, and product segments. In OdinUltra, the "Key Drivers" sub-tab allows investors to explore the critical metrics that support these forecasts. Key drivers are seamlessly identified by our sophisticated models through a comprehensive analysis of ecosystem interconnections that exceeds human capabilities.
To further enhance OdinUltra's offerings, we have incorporated aggregate forecasts for 65 industries within the S&P 500, along with all eleven sectors of the S&P 500. The predictions include aggregate revenue, non-GAAP gross margin, and non-GAAP EBITDA margin. Interestingly, our aggregated fundamentals forecasts carry substantially more significance than those at the individual company level.
OdinUltra applies a scientific methodology to fundamental forecasting, delivering precise, comprehensive, systematic, and unbiased outcomes. We draw from over 1,400 high-quality, publicly available factual data sources, including SEC filings, government agencies, and industry trade groups. We steer clear of sample or panel data, which often suffer from biases due to limited demographic and geographic representation. Even so, relying on a single data source is inherently one-dimensional and can lead to significant errors.
Our algorithms have uncovered more than 400,000 direct and latent relationships between the fundamentals of companies, industries, and economies within the S&P 500. Each company exhibits a unique fingerprint based on its product mix, geographic footprint, and industry-specific metrics. Additionally, modern companies are becoming increasingly complex and interconnected, driven by extensive supply chains, expanding global reach, and increasing diversification in end markets.
For access to our fundamental forecasts, contact us at support@odinultra.ai.
1Total number of companies that Odin has forecasts on key fundamentals metrics.
2Number for Odin’s revenue forecasts that exceeded our high confidence thresholds
3Odin’s average absolute percent error for revenue forecast compared to companies’ actual revenue for high confidence companies.
4Analysts’ average absolute percent error for revenue consensus compared to companies’ actual revenue.
5Accuracy advantage/edge average delivered by Odin compared to analysts’ consensus for revenue in absolute terms.
6Percentage of occurrences when Odin revenue forecasts were above consensus and actual reported revenues were also above consensus (revenue beat).
Published by Big Data Federation · All insights · Subscribe to Insights