Using from Stata
To install via Stata, from the Stata command line run:
ssc install statatwoway
The most efficient way to run StataTwoWay is through Python. Please read From Python on how to install PyTwoWay for Python. A version that does not depend on Python is currently in development.
Sample data: download
To run in Stata:
statatwoway fe cre, config("config.txt") env("stata-env")
To see all available results, type:
return list
The estimators you can run are FE, CRE, or both. Additionally, while a config file is required, an environment is not. However, setting up an environment is recommended. Read about setting up environments in Anaconda here.
Use the following example config file as a template. There is no need to include data or filetype options. StataTwoWay generates a temporary datafile and automatically inputs the name of the generated dataset and filetype dta. For help about all the options that can be included in your config file, from the command line run:
pytw -h
Example config.txt:
col_dict = "{'i': 'your_workerid_col', 'j': 'your_firmid_col', 'y': 'your_compensation_col', 't': 'your_year_col'}"
Note
- Your data must include the following columns:
i
: the worker identifierj
: the firm identifiery
: the compensation variablet
: the time
i |
j |
y |
t |
---|---|---|---|
1 |
1 |
1000 |
2019 |
1 |
2 |
1500 |
2020 |
2 |
3 |
500 |
2019 |
2 |
3 |
550 |
2020 |