How dananghilalkurniawan/smart-money-accumulation-idx works
This repository analyzes seven consecutive days of Indonesian stock-market trading data to identify accumulation and distribution patterns, calculate a Bandar Accumulation Score, and assess its relationship to future returns. The README documents feature engineering, robust normalization, scoring, predictive validation, and visualizations. The tree contains the data files, report, and a notebook, but not the separately described Python modules or figure files; analysis-stage paths therefore point to the existing notebook, and the documented workflow does not confirm every implementation detail.
Main entry points: Code.ipynb Future-return validation · Data Seven-day workbooks · Figures Visualizations
Components 12
Market data
- Seven-day workbooks
Data
Analysis workflow
- Analysis notebook
Src/Code.ipynb - Feature engineering
Src/Code.ipynb - Robust normalization
Src/Code.ipynb - Accumulation score
Src/Code.ipynb - Future-return validation
Src/Code.ipynb - Group comparison
Src/Code.ipynb - Quantile analysis
Src/Code.ipynb
Results
- Visualizations
Figures - Analysis report
Report/laporan_analisis_bandar.pdf
Other components
- IDX data source
- Analyst
Connections 10
- IDX data source → Seven-day workbooks: provides daily data
- Analyst → Analysis notebook: runs analysis
- Analysis notebook → Seven-day workbooks: reads workbooks
- Seven-day workbooks → Feature engineering: supplies trading data
- Feature engineering → Robust normalization: provides features
- Robust normalization → Accumulation score: provides scaled features
- Accumulation score → Future-return validation: tests future returns
- Future-return validation → Group comparison: compares score groups
- Future-return validation → Quantile analysis: evaluates score quantiles
- Future-return validation → Visualizations: supplies results