Qlib ALSTM Configuration for CSI 300 Stock Ranking and Portfolio Backtesting
Summary
This configuration specifies a Qlib experiment that trains an ALSTM model on China’s CSI 300 universe and evaluates stock selections in a portfolio backtest. The data handler uses Alpha360 features, robust feature normalization, missing-value filling, and cross-sectional rank normalization for labels. The prediction target is a short-horizon close-to-close return. Training, validation, and test periods are separated, with the model fitted on historical data before the later test period.
Portfolio construction uses a top-k dropout strategy: it holds up to 50 names and replaces a limited number of positions as rankings change. The backtest includes a benchmark, transaction costs, a minimum fee, closing-price execution, and a market limit threshold. These settings show how model training and portfolio evaluation can be assembled, but the document contains no reported performance or validation results. The stated time ranges and execution assumptions constrain interpretation; realistic conclusions would depend on data quality, point-in-time availability, and whether the modeled costs and fills match actual trading.
Key ideas
- The setup applies an ALSTM model to rank CSI 300 stocks using Alpha360 features.
- Feature normalization and cross-sectional label ranking are included in data preprocessing.
- Training and validation periods precede a distinct test interval.
- The portfolio strategy holds a ranked set of stocks and periodically drops and replaces positions.
- Backtest settings account for benchmark comparison, commissions, minimum fees, and price limits, but no results are provided.
Tags
Full text
# workflow_config_alstm_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: ALSTM
module_path: qlib.contrib.model.pytorch_alstm
kwargs:
d_feat: 6
hidden_size: 64
num_layers: 2
dropout: 0.0
n_epochs: 200
lr: 1e-3
early_stop: 20
batch_size: 800
metric: loss
loss: mse
GPU: 0
rnn_type: GRU
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.