Advanced

• Conditional Average Treatment Effects (CATE) with DoWhy and EconML
  • Linear Model
  • EconML methods
  • Works with any EconML method
  • Avoiding retraining the estimator
  • Refuting the estimate
• Mediation analysis with DoWhy: Direct and Indirect Effects
  • Creating a dataset
  • Step 1: Modeling the causal mechanism
  • Step 2: Identifying the natural direct and indirect effects
  • Step 3: Estimation of the effect
  • Step 4: Refutations
• A Simple Example on Creating a Custom Refutation Using User-Defined Outcome Functions
  • Insert Dependencies
  • Create the Dataset
  • Creating the Causal Model
  • Identify the Estimand
  • Estimating the Effect
  • Refuting the Estimate
  • Using a Randomly Generated Outcome
  • Using a Function that Generates the Outcome from the Confounders
• Estimating effect of multiple treatments
  • Linear model
  • More methods
• Iterating over multiple refutation tests
  • Import Dependencies
  • Inspection Parameters
  • Estimator List
  • Refuter List
  • Create the Datasets
  • Inspect Data
  • Create the CausalModels
  • View Models
  • Identify Effect
  • Identified Estimands
  • Estimate Effect
  • Estimate Values
  • Refute Estimate
  • Refutation Values
• Identifying Effect using ID Algorithm
  • Examples
  • Case 1
  • Case 2
  • Case 3
  • Case 4
  • Case 5
  • Case 6