Periodic Rolling Retraining for Market Forecasting Models

Summary

The document introduces a periodically rolling retraining framework for forecasting models. Its central idea is to refresh a model using up-to-date data at intervals so its forecasts can adapt as market conditions change. The default model is linear, and LightGBM is given as an alternative that can be selected through a configuration setting.

The material describes how to launch the framework and switch model types, but offers no details on the forecast target, retraining schedule, data handling, evaluation process, or trading strategy. It presents no comparative results or evidence that rolling retraining improves forecasts. The framework is therefore an implementation starting point; the document alone is insufficient to judge its predictive quality or practical value.

Key ideas

• The framework periodically retrains forecasting models on recent data to respond to changing market dynamics.
• Linear forecasting is the default model type.
• LightGBM is presented as an alternative selected through configuration.
• The document gives no validation results or details about forecast targets and retraining frequency.

Tags

# Introduction


# Introduction

This is the framework of periodically Rolling Retrain (RR) forecasting models. RR adapts to market dynamics by utilizing the up-to-date data periodically.

## Run the Code
Users can try RR by running the following command:
```bash
    python rolling_benchmark.py run
```

The default forecasting models are `Linear`. Users can choose other forecasting models by changing the `model_type` parameter.
For example, users can try `LightGBM` forecasting models by running the following command:
```bash
    python rolling_benchmark.py --conf_path=workflow_config_lightgbm_Alpha158.yaml run

```

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.