Research Article
[Retracted] Beat Wash-Sale Tax with Multigraph Convolutional Neural Networks Based Trading Strategy
Table 10
Summary for 20% expected return weight portfolio trading without limitation.
| 20% Expected return weight | Model | MDD | Return (%) | Counts | SR | Win rate (%) | Loss (106) | Profit (106) | Net volume (106) |
| Baseline | 0.35 | 45.06 | 1 | 1.31 | | | 0.90 | 0.75 | Random Forest | 0.14 | −3.79 | 2297 | −0.42 | 0.53 | 1.37 | −0.08 | −0.52 | XGBoost | 0.16 | −11.72 | 2290 | −1.34 | 0.51 | 1.21 | −0.23 | −0.56 | GRU | 0.09 | 1.05 | 1927 | 0.08 | 0.51 | 1.16 | 0.02 | −0.38 | LSTM | 0.07 | 16.67 | 1155 | 2.23 | 0.51 | 0.65 | 0.33 | 0.00 | Multi-GCN-GRU | 0.03 | 129.97 | 1718 | 9.67 | 0.59 | 0.42 | 2.60 | 1.52 | Multi-GCN-LSTM | 0.03 | 117.89 | 1601 | 11.67 | 0.62 | 1.01 | 2.36 | 1.15 |
|
|