> ## Documentation Index
> Fetch the complete documentation index at: https://nixtlaverse.nixtla.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Expanding

> Compute expanding mean, std, min, max, and quantile

##

### `expanding_mean`

```python theme={null}
expanding_mean(x, skipna=False)
```

Compute the expanding\_mean of the input array.

**Parameters:**

| Name | Type | Description | Default |
| - | - | - | - |
| `x` | <code>np.ndarray</code> | Input array. | *required* |
| `skipna` | <code>bool</code> | If True, exclude NaN values from calculations. When False (default), any NaN value causes the result to be NaN, maintaining backwards compatibility. When True, NaN values are ignored (matching pandas default behavior). | *required* |

**Returns:**

| Type | Description |
| - | - |
| np.ndarray: Array with the expanding statistic | |

**Examples:**

```pycon theme={null}
>>> import numpy as np
>>> x = np.array([1.0, 2.0, np.nan, 4.0, 5.0])
>>> # Default behavior: NaN propagates
>>> expanding_mean(x)
array([1., 1.5, nan, nan, nan])
>>> # With skipna=True: NaN values are excluded
>>> expanding_mean(x, skipna=True)
array([1., 1.5, 1.5, 2.33..., 3.0])
```

### `expanding_std`

```python theme={null}
expanding_std(x, skipna=False)
```

Compute the expanding\_std of the input array.

**Parameters:**

| Name | Type | Description | Default |
| - | - | - | - |
| `x` | <code>np.ndarray</code> | Input array. | *required* |
| `skipna` | <code>bool</code> | If True, exclude NaN values from calculations. When False (default), any NaN value causes the result to be NaN, maintaining backwards compatibility. When True, NaN values are ignored (matching pandas default behavior). | *required* |

**Returns:**

| Type | Description |
| - | - |
| np.ndarray: Array with the expanding statistic | |

**Examples:**

```pycon theme={null}
>>> import numpy as np
>>> x = np.array([1.0, 2.0, np.nan, 4.0, 5.0])
>>> # Default behavior: NaN propagates
>>> expanding_std(x)
array([1., 1.5, nan, nan, nan])
>>> # With skipna=True: NaN values are excluded
>>> expanding_std(x, skipna=True)
array([1., 1.5, 1.5, 2.33..., 3.0])
```

### `expanding_min`

```python theme={null}
expanding_min(x, skipna=False)
```

Compute the expanding\_min of the input array.

**Parameters:**

| Name | Type | Description | Default |
| - | - | - | - |
| `x` | <code>np.ndarray</code> | Input array. | *required* |
| `skipna` | <code>bool</code> | If True, exclude NaN values from calculations. When False (default), any NaN value causes the result to be NaN, maintaining backwards compatibility. When True, NaN values are ignored (matching pandas default behavior). | *required* |

**Returns:**

| Type | Description |
| - | - |
| np.ndarray: Array with the expanding statistic | |

**Examples:**

```pycon theme={null}
>>> import numpy as np
>>> x = np.array([1.0, 2.0, np.nan, 4.0, 5.0])
>>> # Default behavior: NaN propagates
>>> expanding_min(x)
array([1., 1.5, nan, nan, nan])
>>> # With skipna=True: NaN values are excluded
>>> expanding_min(x, skipna=True)
array([1., 1.5, 1.5, 2.33..., 3.0])
```

### `expanding_max`

```python theme={null}
expanding_max(x, skipna=False)
```

Compute the expanding\_max of the input array.

**Parameters:**

| Name | Type | Description | Default |
| - | - | - | - |
| `x` | <code>np.ndarray</code> | Input array. | *required* |
| `skipna` | <code>bool</code> | If True, exclude NaN values from calculations. When False (default), any NaN value causes the result to be NaN, maintaining backwards compatibility. When True, NaN values are ignored (matching pandas default behavior). | *required* |

**Returns:**

| Type | Description |
| - | - |
| np.ndarray: Array with the expanding statistic | |

**Examples:**

```pycon theme={null}
>>> import numpy as np
>>> x = np.array([1.0, 2.0, np.nan, 4.0, 5.0])
>>> # Default behavior: NaN propagates
>>> expanding_max(x)
array([1., 1.5, nan, nan, nan])
>>> # With skipna=True: NaN values are excluded
>>> expanding_max(x, skipna=True)
array([1., 1.5, 1.5, 2.33..., 3.0])
```

### `expanding_quantile`

```python theme={null}
expanding_quantile(x, p, skipna=False)
```

Compute the expanding\_quantile of the input array.

**Parameters:**

| Name | Type | Description | Default |
| - | - | - | - |
| `x` | <code>[ndarray](#numpy.ndarray)</code> | Input array. | *required* |
| `p` | <code>[float](#float)</code> | Quantile to compute. | *required* |
| `skipna` | <code>[bool](#bool)</code> | If True, exclude NaN values from calculations. When False (default), any NaN value causes the result to be NaN, maintaining backwards compatibility. When True, NaN values are ignored (matching pandas default behavior). | <code>False</code> |

**Returns:**

| Type | Description |
| - | - |
| <code>[ndarray](#numpy.ndarray)</code> | np.ndarray: Array with the expanding statistic |

**Examples:**

```pycon theme={null}
>>> import numpy as np
>>> x = np.array([1.0, 2.0, np.nan, 4.0, 5.0])
>>> # Default behavior: NaN propagates
>>> expanding_quantile(x, 0.5)
array([1., 1.5, nan, nan, nan])
>>> # With skipna=True: NaN values are excluded
>>> expanding_quantile(x, 0.5, skipna=True)
array([1., 1.5, 1.5, 2., 2.5])
```


## Related topics

- [Lag transformations | CoreForecast](/coreforecast/lag_transforms.md)
- [Lag transforms](/mlforecast/lag_transforms.html.md)
- [coreforecast](/coreforecast/index.md)
- [Feature engineering | MLForecast](/mlforecast/feature_engineering.html.md)
- [Pooled lag transforms](/mlforecast/docs/how-to-guides/pooled_lag_transforms.html.md)