# {py:mod}`causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity`

```{py:module} causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity
```

```{autodoc2-docstring} causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity
:allowtitles:
```

## Module Contents

### Functions

````{list-table}
:class: autosummary longtable
:align: left

* - {py:obj}`compute_irm_sensitivity_elements <causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.compute_irm_sensitivity_elements>`
  - ```{autodoc2-docstring} causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.compute_irm_sensitivity_elements
    :summary:
    ```
* - {py:obj}`compute_bias_aware_ci <causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.compute_bias_aware_ci>`
  - ```{autodoc2-docstring} causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.compute_bias_aware_ci
    :summary:
    ```
* - {py:obj}`format_bias_aware_summary <causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.format_bias_aware_summary>`
  - ```{autodoc2-docstring} causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.format_bias_aware_summary
    :summary:
    ```
* - {py:obj}`get_sensitivity_summary <causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.get_sensitivity_summary>`
  - ```{autodoc2-docstring} causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.get_sensitivity_summary
    :summary:
    ```
* - {py:obj}`sensitivity_benchmark <causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.sensitivity_benchmark>`
  - ```{autodoc2-docstring} causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.sensitivity_benchmark
    :summary:
    ```
* - {py:obj}`sensitivity_analysis <causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.sensitivity_analysis>`
  - ```{autodoc2-docstring} causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.sensitivity_analysis
    :summary:
    ```
* - {py:obj}`interpret_sensitivity_analysis <causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.interpret_sensitivity_analysis>`
  - ```{autodoc2-docstring} causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.interpret_sensitivity_analysis
    :summary:
    ```
````

### Data

````{list-table}
:class: autosummary longtable
:align: left

* - {py:obj}`__all__ <causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.__all__>`
  - ```{autodoc2-docstring} causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.__all__
    :summary:
    ```
````

### API

````{py:data} __all__
:canonical: causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.__all__
:value: >
   ['sensitivity_analysis', 'sensitivity_benchmark', 'get_sensitivity_summary', 'interpret_sensitivity_...

```{autodoc2-docstring} causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.__all__
```

````

````{py:function} compute_irm_sensitivity_elements(*, model: typing.Any, y: numpy.ndarray, d: numpy.ndarray, g0: numpy.ndarray, g1: numpy.ndarray, m_hat: numpy.ndarray, w: typing.Optional[numpy.ndarray] = None, w_bar: typing.Optional[numpy.ndarray] = None, psi: typing.Optional[numpy.ndarray] = None, inv_m: typing.Optional[numpy.ndarray] = None, inv_1m: typing.Optional[numpy.ndarray] = None, score: typing.Any = 'ATE') -> dict[str, typing.Any]
:canonical: causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.compute_irm_sensitivity_elements

```{autodoc2-docstring} causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.compute_irm_sensitivity_elements
```
````

````{py:function} compute_bias_aware_ci(effect_estimation: typing.Dict[str, typing.Any] | typing.Any, *, r2_y: float, r2_d: float, rho: float = 1.0, H0: float = 0.0, alpha: float = 0.05, use_signed_rr: bool = False) -> typing.Dict[str, typing.Any]
:canonical: causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.compute_bias_aware_ci

```{autodoc2-docstring} causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.compute_bias_aware_ci
```
````

````{py:function} format_bias_aware_summary(res: typing.Dict[str, typing.Any], label: str | None = None) -> str
:canonical: causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.format_bias_aware_summary

```{autodoc2-docstring} causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.format_bias_aware_summary
```
````

````{py:function} get_sensitivity_summary(effect_estimation: typing.Dict[str, typing.Any] | typing.Any, *, label: typing.Optional[str] = None) -> typing.Optional[str]
:canonical: causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.get_sensitivity_summary

```{autodoc2-docstring} causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.get_sensitivity_summary
```
````

````{py:function} sensitivity_benchmark(effect_estimation: typing.Dict[str, typing.Any] | typing.Any, data: causalis.dgp.causaldata.CausalData, benchmarking_set: typing.List[str] | typing.Literal[all], fit_args: typing.Optional[typing.Dict[str, typing.Any]] = None) -> pandas.DataFrame
:canonical: causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.sensitivity_benchmark

```{autodoc2-docstring} causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.sensitivity_benchmark
```
````

````{py:function} sensitivity_analysis(effect_estimation: typing.Dict[str, typing.Any] | typing.Any, *, r2_y: float, r2_d: float, rho: float = 1.0, H0: float = 0.0, alpha: float = 0.05, use_signed_rr: bool = False) -> typing.Dict[str, typing.Any]
:canonical: causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.sensitivity_analysis

```{autodoc2-docstring} causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.sensitivity_analysis
```
````

````{py:function} interpret_sensitivity_analysis(effect_estimation: typing.Dict[str, typing.Any] | typing.Any, *, r2_y: float, r2_d: float, rho: float = 1.0, H0: float = 0.0, alpha: float = 0.05, use_signed_rr: bool = False) -> typing.Dict[str, typing.Any]
:canonical: causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.interpret_sensitivity_analysis

```{autodoc2-docstring} causalis.scenarios.unconfoundedness.refutation.unconfoundedness.sensitivity.interpret_sensitivity_analysis
```
````
