#forecasting
Financial forecasting and runway modelling with AI agents
EnterpriseArena runs 11 LLMs through a 132-month CFO simulation tracking survival, terminal valuation, and book-closing rates. Only Qwen3.5-9B survives 80% of runs; GPT-5.4 and DeepSeek-V3.1 hit 0%. Human experts achieve 100% survival at 5× the terminal value. The critical bottleneck: LLMs skip ledger reconciliation 80% of the time, acting on stale financial state.
InvestorBench: Qwen2.5-72B leads stock trading at 46.15% CR; finance-tuned Palmyra-Fin backfires on equities—size beats domain fine-tuning.
A NeurIPS 2024 Spotlight paper ablates three LLM-based time series forecasting methods — OneFitsAll, Time-LLM, and CALF — and finds that removing the language model improves accuracy in most cases, with up to a 1,383× training speedup. For finance AI applications like Beancount balance prediction, lightweight purpose-built models consistently beat repurposed LLMs.
FinBen оценява 15 LLM върху 36 финансови набора от данни на NeurIPS 2024, като установява, че GPT-4 достига 0.63 точно съвпадение при числови QA и 0.54 при прогнозиране на движения на акции — близо до случайното. Ето какво означават тези числа за изграждането на надежден счетоводен агент върху Beancount ledger.
