This repository manages codes for replicating “Reassessing the Ins and Outs of Unemployment” (Shimer, 2012), and extends the framework using the latest data during the Covid-19 pandemic.
First, install the package by pip
pip install git+https://github.com/caibengbu/ecma33330_proj.git
After it is installed, simply run
python -m reassessing_the_ins_and_outs_of_unemployment --start=197601 --end=202110
This command activates an analysis from January 1976 to October 2021. You can input any interval as long as there is raw data.
You can specify working directory by passing the --dir
argument
python -m reassessing_the_ins_and_outs_of_unemployment --start=197601 --end=202110 --dir=path/to/wd
You can also specify directory with downloaded raw data by passing the --dir_raw
argument
python -m reassessing_the_ins_and_outs_of_unemployment --start=197601 --end=202110 --dir_raw=path/to/raw
This automatically set the working directory to the parent directory of path/to/raw
.
When the raw data is pre-downloaded and one wants to skip checking the completeness of the files, the -q
argument can be added to avoid automated download and checking.
python -m reassessing_the_ins_and_outs_of_unemployment --start=197601 --end=202110 -q
There are two sources of raw CPS Basic Monthly data: NBER.org and Census.gov. Since Census.gov does not support downloading data files older than 1994 but is more up-to-date than NBER.org in terms of newly published data, we download CPS Basic data older than 1994 from NBER and newer ones from Census.
This package is built on and tested on Python 3.6. It is OS independent, tested on MacOS 10.15.7 and Windows 10. In order to retrieve and install the package, pip
is required. No other manual installation is needed.
If executed without error, there would be an output/
folder and a figure/
folder. output/
contains the monthly transitional probability and transitional rate between employment status. figure/
contains the plot in which hypothetical unemployment rate and actual unemployment rate is plotted together for comparison.