A Fourier-trained neural network using positive Gaussian-Laplace mixtures estimates fixed-horizon densities from empirical characteristic functions, with derived L2 error bounds and competitive performance on benchmarks and equity returns.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
A data-driven Fourier-mixture neural-network method for density estimation
A Fourier-trained neural network using positive Gaussian-Laplace mixtures estimates fixed-horizon densities from empirical characteristic functions, with derived L2 error bounds and competitive performance on benchmarks and equity returns.