Qlib Alpha360 Neural Strategy for CSI 500 Stocks
Summary
This configuration defines a Qlib workflow that trains a PyTorch feedforward neural network on Alpha360 features for CSI 500 stocks. It uses a forward close-price return label, robust feature normalization, missing-value filling, and cross-sectional rank normalization of labels. The dataset spans 2008 to 2020, with training through 2014, validation in 2015–2016, and a test segment beginning in 2017.
For portfolio analysis, a TopkDropoutStrategy holds up to 50 stocks and replaces up to five positions at a time. The backtest uses the Shanghai-Shenzhen 500 index as its benchmark, close-price fills, transaction costs, minimum fees, and a price-limit threshold. The file also configures signal, signal-analysis, and portfolio-analysis records. It provides an experiment recipe rather than performance evidence: no returns, risk statistics, or comparison results are included. Outcomes depend on the specified data, model settings, execution assumptions, and backtest implementation.
Key ideas
- The workflow trains a neural network on Alpha360 features for the CSI 500 universe.
- Features are robustly normalized and missing feature values are filled.
- The training, validation, and test periods are separated by date.
- The portfolio strategy maintains up to 50 holdings and allows five replacements per rebalance.
- The backtest includes benchmark, transaction-cost, minimum-fee, and price-limit assumptions.
Tags
Full text
# workflow_config_mlp_Alpha360_csi500.yaml
```yaml
qlib_init:
provider_uri: "~/.qlib/qlib_data/cn_data"
region: cn
market: &market csi500
benchmark: &benchmark SH000905
data_handler_config: &data_handler_config
start_time: 2008-01-01
end_time: 2020-08-01
fit_start_time: 2008-01-01
fit_end_time: 2014-12-31
instruments: *market
infer_processors:
- class: RobustZScoreNorm
kwargs:
fields_group: feature
clip_outlier: true
- class: Fillna
kwargs:
fields_group: feature
learn_processors:
- class: DropnaLabel
- class: CSRankNorm
kwargs:
fields_group: label
label: ["Ref($close, -2) / Ref($close, -1) - 1"]
port_analysis_config: &port_analysis_config
strategy:
class: TopkDropoutStrategy
module_path: qlib.contrib.strategy
kwargs:
signal: <PRED>
topk: 50
n_drop: 5
backtest:
start_time: 2017-01-01
end_time: 2020-08-01
account: 100000000
benchmark: *benchmark
exchange_kwargs:
limit_threshold: 0.095
deal_price: close
open_cost: 0.0005
close_cost: 0.0015
min_cost: 5
task:
model:
class: DNNModelPytorch
module_path: qlib.contrib.model.pytorch_nn
kwargs:
loss: mse
lr: 0.002
optimizer: adam
max_steps: 8000
batch_size: 4096
GPU: 0
pt_model_kwargs:
input_dim: 360
dataset:
class: DatasetH
module_path: qlib.data.dataset
kwargs:
handler:
class: Alpha360
module_path: qlib.contrib.data.handler
kwargs: *data_handler_config
segments:
train: [2008-01-01, 2014-12-31]
valid: [2015-01-01, 2016-12-31]
test: [2017-01-01, 2020-08-01]
record:
- class: SignalRecord
module_path: qlib.workflow.record_temp
kwargs:
model: <MODEL>
dataset: <DATASET>
- class: SigAnaRecord
module_path: qlib.workflow.record_temp
kwargs:
ana_long_short: False
ann_scaler: 252
- class: PortAnaRecord
module_path: qlib.workflow.record_temp
kwargs:
config: *port_analysis_config
```Shown in full with attribution under the source's licence. Licence: MIT
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.