Back to Community

Neural network architectures for financial time series

DADataSciPro
2d ago
2,715 views
36 posts
strategy
deep-learning
neural-networks

What architectures are people finding useful?

My experiments:

  • LSTM/GRU: Decent for univariate series, overfit on multivariate
  • Temporal Fusion Transformer: Best interpretability, strong performance
  • TCN (Temporal Convolutional Networks): Fast training, competitive accuracy
  • N-BEATS: Great for pure forecasting tasks
The key challenge is preventing overfitting. Financial signal-to-noise is extremely low compared to other domains.

35 Replies

16
RIRiskQuant14h ago

I disagree about the GARCH approach. In my experience, realized volatility estimators (like the Rogers-Satchell estimator) outperform parametric models.

11
GRGradientHunter21h ago

Be careful with the Kelly criterion for position sizing. Full Kelly is way too aggressive. I use quarter-Kelly in practice.

32
COCovarianceKid1d ago

Have you considered using PCA to reduce the dimensionality of the feature space? I found that the first 10 components capture 80%+ of the variance.

15
BUBugHunter992d ago

The competition scoring docs could definitely be clearer. I spent 2 hours debugging what turned out to be a normalization issue.

11
GRGradientHunter2d ago
edited

One more thing: the scoring engine uses a held-out test period that you never see. So your validation score is the best you can do.

9
ENEnsembleKing1d ago

Good point about overfitting. My rule of thumb: never trust a backtest with fewer than 500 observations in the out-of-sample period.

20
QUQuantDev421d ago
edited

Be careful with the Kelly criterion for position sizing. Full Kelly is way too aggressive. I use quarter-Kelly in practice.

Post a Reply