PY Econometrics is a module that allows you to perform different kind of econometric analysis, from standard OLS to most advanced models. It is a fresh module and i'm working on it. This mean that you will not find everything right now or you could find some bugs and so on...
By now, you can find all models in linearmodels.py. The test file is to test all functions. CSV files are dataframes that i use to test and make a comparison with others statistic programs.
Remember: For now, the "main file" is the linearmodels.py file.
In order to use this module you will need the following modules:
- Pandas (dataframe manipulation)
- Numpy (matrix operations)
- Copy (make some usefull replication of the dataframe)
- Tabulate (Prints the final summary)
- Scipy (Computing p value)
The OLS linear regression model can be called by using the class ols in the main file. To call this class, you need the following inputs:
- data is the dataframe where you stored your data you want to pass in the ols model.
- x is the list of strings of all explanatory variables you want to use to explain y. I mean the strings of the columns that store your x's variables. It is very important that you pass a list of strings even if your explanatory variable is just one variable.
- y is the string of the column that stores your dependent variable in your dataframe.
- cons is True by default, but if you want to regress without intercept, just declare cons = False .
- method is "non_robust" by default, but if you want to a Heteroskedastcity robust variance covariance matrix, just declare method = 'robust'
-
To summarize the results, just call "your object name" . summary(). For now, it will print just the table with betas, std, t, p value, and confidence interval.
-
Use betas() if you want to obtain beta coefficients.
-
Use fitted() if you want fitted values.
-
Use residuals() if you want residuals.
To initialize the model:
model = lm.ols(data = df, x=['exper','expersq','kidslt6','kidsge6'],y = 'lwage')
To call the informative summary:
model.summary()
And here is the output:
Linear regression
---------------------------------------------------------------
lwage coefficient se t p_value low 95 high 95
------- ------------- ----------- --------- --------- ----------- ------------
exper 0.0456439 0.0141809 3.21869 0 0.0178494 0.0734385
expersq -0.00101304 0.000417998 -2.42356 0.02 -0.00183232 -0.000193768
kidslt6 0.0314494 0.0892375 0.352424 0.72 -0.143456 0.206355
kidsge6 -0.0345768 0.0283234 -1.22079 0.22 -0.0900906 0.020937
cons 0.875164 0.120301 7.27478 0 0.639374 1.11095
---------------------------------------------------------------
The 2SLS regression model can be called by using the class two_sls in the main file. To call this class, you need the following inputs:
-
data is the dataframe where you stored your data you want to pass in the ols model.
-
exogenous is the list of strings of all exogenous explanatory variables you want to use to explain y. It is very important that you pass a list of strings even if your explanatory variable is just one variable.
-
y is the string of the column that stores your dependent variable in your dataframe.
-
endogenous is the list of strings of all endogenous explanatory variables you want to use to explain y. It is very important that you pass a list of strings even if your explanatory variable is just one variable.
-
instruments is the list of strings of all instrument variables to explain the endogenous one. They must not be in exogenous list. It is very important that you pass a list of strings even if your explanatory variable is just one variable.
-
method is "non_robust" by default, but if you want to a Heteroskedastcity robust variance covariance matrix, just declare method = 'robust'
-
To summarize the results, just call "your object name" . summary(). For now, it will print just the table with betas, std, t, p value, and confidence interval.
-
Use betas() if you want to obtain beta coefficients.
-
Use residuals() if you want residuals.
To initialize the model:
model = lm.two_sls(data = df, exogenous=['exper','expersq','kidslt6','kidsge6'],y = 'lwage', endogenous=['educ'], instruments = ['motheduc','fatheduc','huseduc'] )
To call the informative summary:
model.summary()
And here is the output:
2SlS Regression
---------------------------------------------------------------------------------
lwage coefficient se t p_value low 95 high 95
------- ------------- ----------- --------- --------- ----------- ------------
cons -0.125352 0.290822 -0.431028 0.67 -0.695364 0.444659
exper 0.0415194 0.0134866 3.07856 0 0.0150856 0.0679532
expersq -0.000858582 0.000397676 -2.159 0.03 -0.00163803 -7.91367e-05
kidslt6 -0.0309478 0.0850134 -0.364034 0.72 -0.197574 0.135679
kidsge6 -0.0183228 0.026984 -0.679026 0.5 -0.0712115 0.0345658
educ 0.0793775 0.0219792 3.61148 0 0.0362982 0.122457
---------------------------------------------------------------------------------
INSTRUMENTED: ['educ']
INSTRUMENTS: ['exper', 'expersq', 'kidslt6', 'kidsge6', 'motheduc', 'fatheduc', 'huseduc']
The module allows you to perfor hypthotesis testing on the explanatory variables of your regression model. I implemented the Wald Test divided properly to make sure that it has a F distribution. The Wald Test can be used to test joint significance of variables with a proper R matrix. You can also implement your R matrix if you want to test different hyphotesis, but for now you can use just the standard R matrix.
To use this, you have first to initilize the model to test and then pass the model with the list of variables to test.
model = lm.two_sls(data = df, exogenous=['exper','expersq','kidslt6','kidsge6'],y = 'lwage', endogenous=['educ'], instruments = ['motheduc','fatheduc','huseduc'] )
print(lm.Wald_test(model, ['kidslt6','kidsge6']))
The output is:
{'Wald test': array([[0.3063074]]), 'p_value': array([[0.73632229]])}
- Open your colab notebook and launch the kernel.
- In the file folder (the temporary folder where you can upload every csv and other file), upload the linearmodels.py file and your csv or what else.
- Import linearmodels as lm in your notebook (as lm is optional).
- Finally you can now work with it .
THIS IS NOT THE FINAL PROJECT. I WILL ADD STEP BY STEP ALL THE FUNCTIONS THAT YOU NEED FORO YOUR ECONOMETRIC WORK.