Error when Identifying Effects of Causal Model

Question

I am trying to use the CausalModel and Econml libraries in order to determine the effect of a variable on different scenarios displayed in the dataset below :

So firstly, I import the following libraries :

import pandas as pd
import econml
import dowhy
from dowhy import CausalModel

I then use pandas read_csv to import the dataset and call it "df."

After that I define the Causal Model as the following:

model = CausalModel(data=df.fillna(0),
                    treatment='ai_host.disk.write.bytes',
                    outcome='scenario',
                    common_causes='col'
                    )

model.view_model()

With the following being the output

After that I generate the estimand:

identified_estimand= model.identify_effect(proceed_when_unidentifiable=True)
print(identified_estimand)

With the following output:

Estimand type: nonparametric-ate

### Estimand : 1
Estimand name: backdoor
Estimand expression:
             d                                        
───────────────────────────(Expectation(scenario|col))
d[ai_host.disk.write.bytes]                           
Estimand assumption 1, Unconfoundedness: If U→{ai_host.disk.write.bytes} and U→scenario then P(scenario|ai_host.disk.write.bytes,col,U) = P(scenario|ai_host.disk.write.bytes,col)

### Estimand : 2
Estimand name: iv
No such variable found!

### Estimand : 3
Estimand name: frontdoor
No such variable found!

After this I finally try calculate the Causal Effect :

identified_estimand_experiment = model.identify_effect(proceed_when_unidentifiable=True)

from sklearn.ensemble import RandomForestRegressor
metalearner_estimate = model.estimate_effect(identified_estimand_experiment,
method_name="backdoor.econml.metalearners.TLearner",
confidence_intervals=False,
method_params={
     "init_params":{'models': RandomForestRegressor()},
     "fit_params":{}
              })
print(metalearner_estimate)

But I get the following error each time :

ValueError                                Traceback (most recent call last)
<ipython-input-15-6f34377dbe77> in <module>()
      8 method_params={
      9      "init_params":{'models': RandomForestRegressor()},
---> 10      "fit_params":{}
     11               })
     12 print(metalearner_estimate)

7 frames
/usr/local/lib/python3.7/dist-packages/sklearn/preprocessing/_encoders.py in _transform(self, X, handle_unknown, force_all_finite, warn_on_unknown)
    140                         " during transform".format(diff, i)
    141                     )
--> 142                     raise ValueError(msg)
    143                 else:
    144                     if warn_on_unknown:

ValueError: Found unknown categories [0] in column 0 during transform

Please may someone assist me and understanding and rectifying this error. Please also note that in order to use Econml, you need Python 3.8 and lower.

Answer 1

I was also encountering this problem but when I used a linear regression model instead of the Random Forest Regressor metalearner I had no issues.

This requires replacing

identified_estimand_experiment = model.identify_effect(proceed_when_unidentifiable=True)

from sklearn.ensemble import RandomForestRegressor
metalearner_estimate = 
model.estimate_effect(identified_estimand_experiment,
method_name="backdoor.econml.metalearners.TLearner",
confidence_intervals=False,
method_params={
               "init_params":{'models': RandomForestRegressor()},
               "fit_params":{}
               })
print(metalearner_estimate)

    

with

identified_estimand_experiment = model.identify_effect(proceed_when_unidentifiable=True)

linreg_estimate = model.estimate_effect(identified_estimand_experiment,
                            method_name="backdoor.linear_regression",
                            confidence_intervals=False)
print(linreg_estimate)

Other methods like using

method_name = "backdoor.propensity_score_stratification" 

and

method_name = "backdoor.propensity_score_matching"

might also be of interest.