causal_agent.methods.regression_discontinuity package

Submodules

causal_agent.methods.regression_discontinuity.diagnostics module

Diagnostic checks for Regression Discontinuity Design (RDD).

causal_agent.methods.regression_discontinuity.diagnostics.run_rdd_diagnostics(df, outcome, running_variable, cutoff, covariates=None, bandwidth=None)[source]

Runs diagnostic checks for RDD analysis.

Currently includes: - Covariate Balance Check (t-tests) Placeholders for: - Density Test (McCrary) - Placebo Cutoff Tests - Bandwidth Sensitivity

Parameters:

  • df (DataFrame) – Input DataFrame.

  • outcome (str) – Name of the outcome variable.

  • running_variable (str) – Name of the running variable.

  • cutoff (float) – The threshold value.

  • covariates (List[str] | None) – Optional list of covariate names to check for balance.

  • bandwidth (float | None) – Optional bandwidth to restrict the analysis. If None, a default is used.

Returns:

Dictionary containing diagnostic results.

Return type:

Dict[str, Any]

causal_agent.methods.regression_discontinuity.estimator module

Regression Discontinuity Design (RDD) Estimator.

Tries to use DoWhy’s RDD implementation first, falling back to a basic comparison of linear fits around the cutoff if DoWhy fails.

causal_agent.methods.regression_discontinuity.estimator.estimate_effect_dowhy(df, treatment, outcome, running_variable, cutoff_value, covariates, **kwargs)[source]

Estimate RDD effect using DoWhy.

causal_agent.methods.regression_discontinuity.estimator.estimate_effect_fallback(df, treatment, outcome, running_variable, cutoff_value, covariates, **kwargs)[source]

Estimate RDD effect using simple linear regression comparison fallback.

causal_agent.methods.regression_discontinuity.estimator.effect_estimate_rdd(df, outcome, running_variable, cutoff_value, treatment=None, covariates=None, bandwidth=None, **kwargs)[source]

Estimates RDD effect using the ‘evan-magnusson/rdd’ package. Uses IK optimal bandwidth selection from the same package by default.

causal_agent.methods.regression_discontinuity.estimator.estimate_effect(df, treatment, outcome, running_variable, cutoff_value, covariates=None, bandwidth=None, query=None, llm=None, **kwargs)[source]

Estimates the causal effect using Regression Discontinuity Design.

Tries DoWhy implementation first if use_dowhy=True, otherwise uses fallback.

Parameters:

  • df (DataFrame) – Input DataFrame.

  • treatment (str) – Name of the treatment variable (often implicitly defined by cutoff). DoWhy might still need it, fallback doesn’t use it directly.

  • outcome (str) – Name of the outcome variable.

  • running_variable (str) – Name of the variable determining treatment assignment.

  • cutoff – The threshold value for the running variable.

  • covariates (List[str] | None) – Optional list of covariate names (support varies).

  • bandwidth (float | None) – Optional bandwidth around the cutoff. If None, a default is used.

  • use_dowhy – Whether to attempt using the DoWhy library first.

  • query (str | None) – Optional user query for context.

  • llm (langchain.chat_models.base.BaseChatModel | None) – Optional Language Model instance.

  • **kwargs – Additional keyword arguments for underlying methods.

Returns:

Dictionary containing estimation results.

Return type:

Dict[str, Any]

causal_agent.methods.regression_discontinuity.llm_assist module

LLM assistance functions for Regression Discontinuity Design (RDD).

causal_agent.methods.regression_discontinuity.llm_assist.suggest_rdd_parameters(df_cols, query, llm=None)[source]

(Placeholder) Use LLM to suggest RDD parameters (running variable, cutoff).

Parameters:

  • df_cols (List[str]) – List of available column names.

  • query (str) – User’s causal query text.

  • llm (langchain.chat_models.base.BaseChatModel | None) – Optional LLM model instance.

Returns:

Dictionary containing suggested ‘running_variable’ and ‘cutoff’, or empty.

Return type:

Dict[str, Any]

causal_agent.methods.regression_discontinuity.llm_assist.interpret_rdd_results(results, diagnostics, llm=None)[source]

Use LLM to interpret Regression Discontinuity Design (RDD) results.

Parameters:

  • results (Dict[str, Any]) – Dictionary of estimation results from the RDD estimator.

  • diagnostics (Dict[str, Any] | None) – Dictionary of diagnostic test results.

  • llm (langchain.chat_models.base.BaseChatModel | None) – Optional LLM model instance.

Returns:

String containing natural language interpretation.

Return type:

str