Table 2 Comparison results of stock indicators between StockMixer with ATFNet and existing stock prediction methods on the US stock dataset.

From: Research on deep learning model for stock prediction by integrating frequency domain and time series features

Model

NASDAQ

NYSE

 

IC

RIC

Prec@N

SR

IC

RIC

Prec@N

SR

Linear

0.019

0.188

0.505

0.517

0.015

0.163

0.497

0.625

LSTM

0.032

0.354

0.514

0.892

0.024

0.256

0.512

0.857

ALSTM

0.035

0.371

0.522

0.941

0.023

0.276

0.519

0.764

RGCN

0.034

0.382

0.516

1.054

0.025

0.275

0.517

0.932

GAT

0.035

0.377

0.530

1.233

0.025

0.297

0.521

1.070

RSR-I

0.038

0.398

0.531

1.238

0.026

0.284

0.519

0.998

STHAN-SR

0.039

0.451

0.543

1.416

0.029

0.344

0.542

1.228

ESTIMATE

0.037

0.444

0.539

1.307

0.030

0.327

0.536

1.115

Our

0.041

0.473

0.577

1.333

0.028

0.347

0.557

1.233

  1. Bold values indicate the best performance achieved by the proposed StockMixer with ATFNet model for the corresponding evaluation metrics.
Back to article page