Project File Management

I suggest using a numbering system to manage project files. For example, the file structure of the 2022_network_analysis project would be as follows:

2022_network_analysis
  \-01_code
    \-a01_analysis.ipynb
    \-a02_visualize.ipynb
    \-d11_basic_data.ipynb
  \-02_rdata
    \-raw_data.csv
  \-03_wdata
    \-0100_reg.csv
    \-0200_visual.csv
    \-1100_data_wip.csv
    \-1101_data_processed.csv
  \-04_result
    \-tab_baseline.rtf
    \-fig_trend.svg

Code¶

The code is stored in 01_code.

• a01 represents the first code used for analysis (taking the first letter of "analysis").
• d11 represents the processing (taking the first letter of "data") of the 11 series of data.

Raw Data¶

The raw data is stored in 02_rdata, which includes externally obtained raw data such as raw_data.csv.

Try to avoid changing the original names of the raw data (unless they are files with unclear names exported from platforms like WRDS or CSMAR) to facilitate locating the source of each piece of raw data in the future.

Processed Data¶

The processed data is stored in 03_wdata.

• The components of the data can be written with the prefix 11 or 12.
  • They can be classified by data source, such as using the 11 prefix for Compustat data and the 12 prefix for CRSP data.
  • They can also be classified by topic, such as using the 11 prefix for stock price data and the 12 prefix for macroeconomic data.
• Data used for analysis starts with 01-09, such as 01 for regression analysis data and 02 for visualization data.
• The last two digits can be sequentially numbered from 00 to 99.

Results¶

The output results are stored in 04_result.

• The prefix tab is used for tables.
• The prefix fig is used for images.