Research Article
[Retracted] Beat Wash-Sale Tax with Multigraph Convolutional Neural Networks Based Trading Strategy
Table 16
Summary for 40% expected return weight portfolio trading without limitation.
| 40% Expected return weight | Model | MDD | Return (%) | Counts | SR | Win rate (%) | Loss (106) | Profit (106) | Net volume (106) |
| Baseline | 0.37 | 163.91 | 1 | 2.46 | | | 3.28 | 2.65 | Random Forest | 0.18 | 91.56 | 1398 | 2.40 | 0.33 | 2.39 | 1.83 | 0.31 | XGBoost | 0.17 | 124.15 | 1422 | 3.08 | 0.33 | 1.83 | 2.48 | 0.92 | GRU | 0.24 | 17.69 | 1187 | 0.69 | 0.33 | 1.57 | 0.35 | −0.32 | LSTM | 0.12 | 25.55 | 680 | 1.38 | 0.29 | 1.34 | 0.51 | −0.14 | Multi-GCN-GRU | 0.06 | 142.52 | 1097 | 5.49 | 0.43 | 0.28 | 2.85 | 1.73 | Multi-GCN-LSTM | 0.06 | 252.94 | 960 | 6.70 | 0.45 | 0.66 | 5.06 | 2.98 |
|
|