Configuring an Alpha360 SFM Stock Ranking Backtest
Summary
This Qlib configuration defines a Chinese equity workflow using the CSI 300 universe and its associated benchmark. It prepares Alpha360 features with robust score normalization and missing-value filling, while labels are rank-normalized after missing labels are dropped. The target is a forward close-to-close return. The data is divided into training, validation, and test periods, with the model fit on the training period.
The model is Qlib's SFM implementation, configured with a mean squared error loss and Adam optimizer. The portfolio layer uses a top-k dropout strategy that holds a selected group and replaces a smaller number of names as signals change. The backtest specifies close-price execution, transaction costs, a minimum fee, and a price-limit threshold, and records signal and portfolio analyses. This is an experiment specification, not evidence of profitability: it gives no reported metrics, and results depend on the supplied dataset, model outputs, and assumptions about execution and costs.
Key ideas
- The workflow applies Qlib's Alpha360 handler to the CSI 300 universe.
- Feature normalization and missing-value processing are specified separately from label processing.
- The forward return label and time segments define the prediction task and evaluation periods.
- An SFM model is trained with Adam and mean squared error loss.
- The portfolio backtest specifies top-k turnover behavior, close-price fills, fees, and a price-limit threshold.
Tags
Full text
# workflow_config_sfm_Alpha360.yaml
```yaml
qlib_init:
provider_uri: "~/.qlib/qlib_data/cn_data"
region: cn
market: &market csi300
benchmark: &benchmark SH000300
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: SFM
module_path: qlib.contrib.model.pytorch_sfm
kwargs:
d_feat: 6
hidden_size: 64
output_dim: 32
freq_dim: 25
dropout_W: 0.5
dropout_U: 0.5
n_epochs: 20
lr: 1e-3
batch_size: 1600
early_stop: 20
eval_steps: 5
loss: mse
optimizer: adam
GPU: 0
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.