How can I calculate the treatment effect function in a double machine learning model?
Wangpc-972 opened this issue · 1 comments
Wangpc-972 commented
On this website, the fourth line of code is:
# Treatment effect function
def exp_te(x):
return np.exp(2*x[0])
How can I obtain such an equation from real data? Or how can I calculate the 'True effect'?
Looking forward to your reply.
Thank you/
kbattocchi commented
In a real application, the true treatment effect will be unknown. The effect method of our estimators is how you'd estimate it; for example the notebook you're referencing shows the estimates of a few different estimators compared to the true effect after cell 10.