# Types

The public functions in `idid` use a small set of type aliases and typed-dict
configuration objects. They are documented here directly instead of being
re-exported from the top-level package.

```{py:currentmodule} idid._types
```

## Type Aliases

```{py:data} Control
:canonical: idid._types.Control

Control-group rule accepted by {py:func}`idid.estimate`.

Allowed values:

- `"never"`: use the never-exposed cohort, coded as `E = 0`.
- `"notyet"`: use units not yet exposed by time `t` as controls.
```

```{py:data} Method
:canonical: idid._types.Method

Cell-level estimation method accepted by {py:func}`idid.estimate`.

Allowed values:

- `"dr"`: doubly robust estimation.
- `"dml"`: double machine learning.
- `"reg"`: outcome regression.
- `"ipw"`: inverse-probability weighting.
- `"std_ipw"`: standardized inverse-probability weighting.
```

```{py:data} AggMethod
:canonical: idid._types.AggMethod

Aggregation method accepted by {py:func}`idid.agg_latt`.

Allowed values:

- `"simple"`: aggregate all post-exposure LATTs.
- `"group"` / `"cohort"`: aggregate by exposure cohort.
- `"cohort_custom"`: aggregate using a user-supplied cohort grouping.
- `"calendar"`: aggregate by calendar time.
- `"calendar_cumu"`: cumulative calendar-time aggregation.
- `"dynamic"`: aggregate by event-time horizon.
- `"dynamic_bal"`: balanced dynamic aggregation.
```

```{py:data} FloatArray
:canonical: idid._types.FloatArray

Floating-point NumPy array type used in lower-level estimator and nuisance
model interfaces.
```

## Configuration Objects

`````{py:class} OutcomeModel()
:canonical: idid._types.OutcomeModel

Protocol for outcome nuisance models used by the DML estimators.

Required methods:

- `fit(X, y)`
- `predict(X)`
`````

`````{py:class} TreatmentModel()
:canonical: idid._types.TreatmentModel

Protocol for treatment / propensity nuisance models used by the DML
estimators.

Required methods:

- `fit(X, y)`
- `predict_proba(X)`
`````

`````{py:class} OutcomeRegressionKwargs()
:canonical: idid._types.OutcomeRegressionKwargs

Bases: {py:obj}`typing.TypedDict`

```{autodoc2-docstring} idid._types.OutcomeRegressionKwargs
```

````{py:attribute} ns_est
:canonical: idid._types.OutcomeRegressionKwargs.ns_est
:type: idid._types.NuisanceEstimator

```{autodoc2-docstring} idid._types.OutcomeRegressionKwargs.ns_est
```
````

````{py:attribute} use_levels
:canonical: idid._types.OutcomeRegressionKwargs.use_levels
:type: bool

```{autodoc2-docstring} idid._types.OutcomeRegressionKwargs.use_levels
```
````
`````

`````{py:class} DMLKwargs()
:canonical: idid._types.DMLKwargs

Bases: {py:obj}`typing.TypedDict`

```{autodoc2-docstring} idid._types.DMLKwargs
```

````{py:attribute} nfolds
:canonical: idid._types.DMLKwargs.nfolds
:type: int

```{autodoc2-docstring} idid._types.DMLKwargs.nfolds
```
````

````{py:attribute} m_m
:canonical: idid._types.DMLKwargs.m_m
:type: idid._types.OutcomeModel

```{autodoc2-docstring} idid._types.DMLKwargs.m_m
```
````

````{py:attribute} g_m
:canonical: idid._types.DMLKwargs.g_m
:type: idid._types.TreatmentModel

```{autodoc2-docstring} idid._types.DMLKwargs.g_m
```
````

````{py:attribute} p_m
:canonical: idid._types.DMLKwargs.p_m
:type: idid._types.TreatmentModel

```{autodoc2-docstring} idid._types.DMLKwargs.p_m
```
````
`````