Paper: arXiv 2609.26303

Authors: Masoud Soleimani

Abstract

Forecasters often score the same units per date against one standardized realized outcome. We show that every standardized forecast splits exactly into a component aligned with this common target and a component uncorrelated with it. Three consequences follow: forecast-error correlation largely mirrors forecast correlation and is therefore a poor measure of diversity; an equally weighted combination beats a no-information forecast only when average alignment is large relative to the combination’s dispersion; and the gain from adding a forecaster separates into genuine improvement and mere dilution, which equal-weight admission can mistakenly reward. We develop a cautious selection rule, study it in simulations, and apply it to language-model forecasts of US equity rankings and mechanical signals ranking exchange-traded funds. Selection removes most dilution losses, but no combination beats the no-information forecast.

Complexity vs Empirical Score

  • Math Complexity: 8.5/10
  • Empirical Rigor: 7.5/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: This paper presents a highly mathematical decomposition of forecast alignment and its implications for combination, backed by simulations and real-world applications. The theoretical contributions are significant, and the empirical sections provide solid validation, even if the ’null findings’ are somewhat counterintuitive.

Research Flowchart

  flowchart TD
    A[Research Goal: Understand cross-sectional forecast dynamics] --> B{Key Concepts: Target alignment, Dilution, Forecast selection};
    B --> C[Methodology: Decompose standardized forecasts into aligned/uncorrelated components];
    C --> D{Data/Inputs: Language model forecasts (US equity), Mechanical signals (ETFs)};
    D --> E[Computational Processes: Analyze forecast-error correlation, Simulate combination strategies, Apply cautious selection rule];
    E --> F[Key Findings: Forecast-error correlation mirrors forecast correlation; Equal-weight combination effectiveness depends on alignment/dispersion; Cautious selection removes dilution losses but no combination beats no-information forecast.];

Cited References (60)

  1. The combination of forecasts (1969) — link
  2. Diversity creation methods: A survey and categorisation (2005) — link
  3. A nonparametric approach to identifying a subset of forecasters that outperforms the simple average (2017) — link
  4. Econometric evaluation of linear macro-economic models (1986) — link
  5. The forecast combination puzzle: A simple theoretical explanation (2016) — link
  6. Combining forecasts: A review and annotated bibliography (1989) — link
  7. Comparing predictive accuracy (1995) — link
  8. Machine learning for regularized survey forecast combination: Partially-egalitarian LASSO and its derivatives (2019) — link
and 52 more
  1. Combining expert forecasts: Can anything beat the simple average? (2013) — link
  2. Improved methods of combining forecasts (1984) — link
  3. Tests for forecast encompassing (1998) — link
  4. Beyond forecast leaderboards: Measuring individual model importance based on contribution to ensemble accuracy (2026) — link
  5. Neural network ensembles, cross validation, and active learning (1995)
  6. A well-conditioned estimator for large-dimensional covariance matrices (2004) — link
  7. Improving the simple average combined forecast via factor-adjusted regularization (2026) — link
  8. Combining forecasts under structural breaks using graphical LASSO (2026) — link
  9. On the uncertainty of a combined forecast: The critical role of correlation (2023) — link
  10. Optimal selection of expert forecasts with integer programming (2018) — link
  11. The evaluation of economic forecasts (1969)
  12. A simple, positive semi-definite, heteroskedasticity and autocorrelation consistent covariance matrix (1987) — link
  13. Automatic lag selection in covariance matrix estimation (1994) — link
  14. Too similar to combine? O n negative weights in forecast combination (2023) — link
  15. A simple explanation of the forecast combination puzzle (2009) — link
  16. Combination forecasts of output growth in a seven-country data set (2004) — link
  17. Forecast combinations (2006) — link
  18. Generalization error of ensemble estimators (1996) — link
  19. Forecast combinations: An over 50-year review (2023) — link
  20. Persistence in forecasting performance and conditional combination strategies (2006) — link
  21. The cross-section of volatility and expected returns (2006) — link
  22. Value and momentum everywhere (2013) — link
  23. Forecaster diversity and the benefits of combining forecasts (1995) — link
  24. Valid post-selection inference (2013) — link
  25. Survivorship bias in performance studies (1992) — link
  26. Identifying expertise to extract the wisdom of crowds (2015) — link
  27. Forecast combination with entry and exit of experts (2009) — link
  28. Limits for the precision and value of information from dependent sources (1985) — link
  29. The use of confidence or fiducial limits illustrated in the case of the binomial (1934) — link
  30. Optimal combination of survey forecasts (2015) — link
  31. Tests of conditional predictive ability (2006) — link
  32. Assessing look-ahead bias in stock return predictions generated by GPT sentiment analysis — → 2309.17322
  33. Making and evaluating point forecasts (2011) — link
  34. The model confidence set (2011) — link
  35. Tests for multiple forecast encompassing (2000)
  36. Pooling of forecasts (2004) — link
  37. Computing the nearest correlation matrix— A problem from finance (2002) — link
  38. A simple sequentially rejective multiple test procedure (1979) — link
  39. Simultaneous inference in general parametric models (2008) — link
  40. Evidence of predictable behavior of security returns (1990) — link
  41. Returns to buying winners and selling losers: Implications for stock market efficiency (1993) — link
  42. The jackknife and the bootstrap for general stationary observations (1989) — link
  43. Honey, I shrunk the sample covariance matrix (2004) — link
  44. Economic forecast evaluation: Profits versus the conventional error measures (1991)
  45. Why do some combinations perform better than others? (2020) — link
  46. Optimal and robust combination of forecasts via constrained optimization and shrinkage (2022) — link
  47. Stepwise multiple testing as formalized data snooping (2005) — link
  48. Wisdom of the silicon crowd: LLM ensemble prediction capabilities rival human crowd accuracy (2024) — link
  49. A comparison of the two one-sided tests procedure and the power approach for assessing the equivalence of average bioavailability (1987) — link
  50. Out-of-sample tests of forecasting accuracy: An analysis and review (2000) — link
  51. Regression shrinkage and selection via the lasso (1996) — link
  52. A unified theory of diversity in ensemble learning (2023) — link

1 of 60 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.