Newest papers

Target alignment, dilution and forecast selection when cross-sectional forecasts share a common target

Masoud Soleimani · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2609.26303 · cited by 0 works (OpenAlex) · open-access PDF

Indonesian

Prediktor sering mencetak unit yang sama per tanggal terhadap satu hasil yang terstandarisasi. Kami menunjukkan bahwa setiap prediksi yang terstandarisasi terpecah persis menjadi komponen yang sejalan dengan target umum ini dan komponen yang tidak terkait dengannya. Tiga konsekuensi mengikuti: korelasi prediksi-kesalahan sebagian besar mencerminkan korelasi prediksi dan karena itu merupakan ukuran yang buruk dari keragaman; kombinasi yang seimbang mengalahkan prediksi non-informasi hanya ketika persamaan rata-rata besar dibandingkan dengan dispersi kombinasi; dan keuntungan dari menambahkan prediktor memisahkan perbaikan sejati dan pelembaban sederhana, yang dapat disesuaikan dengan kesalahan. Kami mengembangkan aturan pemilihan yang hati-hati, mempelajarinya dalam simulasi, dan menerapkannya prediksi model bahasa dari peringkat ekuitas AS dan sinyal mekanis peringkat dana pertukaran.

Machine translation (Meta M2M100); the English original follows.

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.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Selection (genetic algorithm), Uncorrelated, Econometrics, Statistics, Ranking (information retrieval), Measure (data warehouse)

Is this paper real? This page was built from OpenAlex and arXiv records, not generated. Software can verify any reference in one call: GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2609.26303, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txt

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange