Qlib LightGBM Pipeline for CSI 300 Stock Ranking
Summary
This Qlib configuration defines a daily stock-ranking experiment for the CSI 300 universe, using the CSI 300 index as its benchmark. Its features combine price and volume transformations, including residual and fit measures, intraday range, and rolling correlations between returns and volume changes. The label is a forward close-to-close return, and cross-sectional z-score normalization is applied to labels after missing labels are dropped.
A LightGBM gradient-boosted model is trained on an earlier historical segment, validated on a subsequent period, and evaluated on a later test segment. The portfolio layer uses a top-k dropout strategy: it holds a selected group of predictions and replaces only some names at a time. The backtest specifies closing-price execution, transaction costs, a minimum fee, and a price limit threshold, with signal and portfolio analysis records enabled. This is an experiment specification, not a report of results; it does not show returns, robustness checks, or whether the feature and execution assumptions avoid leakage or match live trading.
Key ideas
- The experiment ranks CSI 300 stocks using daily price and volume features.
- Its target is a forward close-to-close return and labels receive cross-sectional normalization.
- LightGBM is trained, validated, and tested on sequential historical date segments.
- The portfolio strategy holds top-ranked names and periodically replaces a subset.
- Backtest settings include a benchmark, closing-price fills, transaction costs, and price limits.
- The configuration provides no evidence of realized performance or live-trading robustness.
Tags
Full text
# workflow_config_lightgbm_configurable_dataset.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
instruments: *market
data_loader:
class: QlibDataLoader
kwargs:
config:
feature:
- ["Resi($close, 15)/$close", "Std(Abs($close/Ref($close, 1)-1)*$volume, 5)/(Mean(Abs($close/Ref($close, 1)-1)*$volume, 5)+1e-12)", "Rsquare($close, 5)", "($high-$low)/$open", "Rsquare($close, 10)", "Corr($close, Log($volume+1), 5)", "Corr($close/Ref($close,1), Log($volume/Ref($volume, 1)+1), 5)", "Corr($close, Log($volume+1), 10)", "Rsquare($close, 20)", "Corr($close/Ref($close,1), Log($volume/Ref($volume, 1)+1), 60)", "Corr($close/Ref($close,1), Log($volume/Ref($volume, 1)+1), 10)", "Corr($close, Log($volume+1), 20)", "(Less($open, $close)-$low)/$open"]
- ["RESI5", "WVMA5", "RSQR5", "KLEN", "RSQR10", "CORR5", "CORD5", "CORR10", "RSQR20", "CORD60", "CORD10", "CORR20", "KLOW"]
label:
- ["Ref($close, -2)/Ref($close, -1) - 1"]
- ["LABEL0"]
freq: day
learn_processors:
- class: DropnaLabel
- class: CSZScoreNorm
kwargs:
fields_group: label
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: LGBModel
module_path: qlib.contrib.model.gbdt
kwargs:
loss: mse
colsample_bytree: 0.8879
learning_rate: 0.2
subsample: 0.8789
lambda_l1: 205.6999
lambda_l2: 580.9768
max_depth: 8
num_leaves: 210
num_threads: 20
dataset:
class: DatasetH
module_path: qlib.data.dataset
kwargs:
handler:
class: DataHandlerLP
module_path: qlib.data.dataset.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.