Replication Package for Anonymous Journal Submission

Description

The package provides code, processed inputs, generated tables and figures, and documentation for reproducing the aggregate portfolio-exposure measures and risk-aversion calibration targets reported in the paper. The analysis combines public Survey of Consumer Finances portfolio data with public Fama-French market-return data and Shiller market-state data. Raw public data sources are documented in raw_data/raw_data_manifest.csv. The processed files in input/ serve as a lightweight production cache, while code/rebuild_from_raw.py and code/rebuild_outputs.py document the rebuild workflows. The dataset is intended to support transparency and reproducibility; it does not contain proprietary data.

Steps to reproduce

The dataset contains a prepared replication package. The file raw_data/raw_data_manifest.csv lists the public raw data sources used in the analysis, including Survey of Consumer Finances files, Fama-French market-return data, and Shiller market-state data. The processed files in input/ provide a lightweight production cache for rebuilding the article tables and figure. To reproduce the article outputs from the cached processed inputs, run: python code/rebuild_outputs.py To rebuild the analysis from raw public data, place the raw files listed in raw_data/raw_data_manifest.csv in the locations described in raw_data/README.md, then run: python code/rebuild_from_raw.py The scripts regenerate the processed inputs, table fragments, summary CSV files, and figure files used in the article. Outputs are written to output/.

Institutions

• Lund University

  Skåne, Lund

Categories

Licence