A decision-focused neural network using GPT-2 embeddings for stocks and macro context reports higher Sharpe ratios and terminal wealth than deep learning baselines, but flawed proofs and survivorship-biased data undermine the claim.
In this formulation, ˆΣ−1 t+h depends on ˆLt+h through the relation ˆΣt+h = ˆLt+h ˆL⊤ t+h
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
q-fin.PM 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Decision-informed Neural Networks with Large Language Model Integration for Portfolio Optimization
A decision-focused neural network using GPT-2 embeddings for stocks and macro context reports higher Sharpe ratios and terminal wealth than deep learning baselines, but flawed proofs and survivorship-biased data undermine the claim.