pystatpower.models.proportion.single.ci ¶

Functions:

Name	Description
`solve_distance`	Calculate the confidence interval width for one proportion or the distance from the proportion to the confidence bound.
`solve_size`	Estimate the required sample size, given either a desired confidence interval width or a desired distance from the proportion to the confidence bound.
`solve_proportion`	Estimate the required proportion, given either a desired confidence interval width or a desired distance from the proportion to the confidence bound.

solve_distance ¶

solve_distance(
    *,
    proportion: float,
    size: int,
    conf_level: float = 0.95,
    interval_type: Literal[
        "two-sided", "lower one-sided", "upper one-sided"
    ] = "two-sided",
    method: Literal[
        "clopper-pearson", "wald", "wilson"
    ] = "clopper-pearson",
    continuity_correction: bool = False,
) -> float

Calculate the confidence interval width for one proportion or the distance from the proportion to the confidence bound.

Parameters:

Name	Type	Description	Default
`proportion`	`float`	Proportion.	required
`size`	`int`	Sample size.	required
`conf_level`	`float`	Confidence level.	`0.95`
`interval_type`	`Literal['two-sided', 'lower one-sided', 'upper one-sided']`	Type of the confidence interval. `two-sided`: Two-sided confidence interval \((L, U)\). `lower one-sided`: Lower one-sided confidence interval \((L, +\infty)\). `upper one-sided`: Upper one-sided confidence interval \((-\infty, U)\).	`'two-sided'`
`method`	`Literal['clopper-pearson', 'wald', 'wilson']`	Method used in calculation of the confidence interval.	`'clopper-pearson'`
`continuity_correction`	`bool`	Whether or not to apply Yate's continuity correction, only valid for `method='wald'` or `method='wilson'`.	`False`

Returns:

Type	Description
`float`	The confidence interval width.

Raises:

Type	Description
`ValueError`	If `method='wald'` or `method='wilson'`, and `continuity_correction=None`.

solve_size ¶

solve_size(
    *,
    proportion: float,
    distance: float,
    conf_level: float = 0.95,
    interval_type: Literal[
        "two-sided", "lower one-sided", "upper one-sided"
    ] = "two-sided",
    method: Literal[
        "clopper-pearson", "wald", "wilson"
    ] = "clopper-pearson",
    continuity_correction: bool = False,
) -> int

Estimate the required sample size, given either a desired confidence interval width or a desired distance from the proportion to the confidence bound.

Parameters:

Name	Type	Description	Default
`proportion`	`float`	Proportion.	required
`distance`	`float`	if `interval_type='two-sided'`, provide confidence interval width. if `interval_type='lower one-sided'` or `interval_type='upper one-sided'`, provide the distance from the proportion to the confidence bound.	required
`conf_level`	`float`	Confidence level.	`0.95`
`interval_type`	`Literal['two-sided', 'lower one-sided', 'upper one-sided']`	Type of the confidence interval. `two-sided`: Two-sided confidence interval \((L, U)\). `lower one-sided`: Lower one-sided confidence interval \((L, +\infty)\). `upper one-sided`: Upper one-sided confidence interval \((-\infty, U)\).	`'two-sided'`
`method`	`Literal['clopper-pearson', 'wald', 'wilson']`	Method used in calculation of the confidence interval.	`'clopper-pearson'`
`continuity_correction`	`bool \| None`	Whether or not to apply Yate's continuity correction, only valid for `method='wald'` or `method='wilson'`.	`False`

Returns:

Type	Description
`int`	The required sample size.

Raises:

Type	Description
`ValueError`	If `method='wald'` or `method='wilson'`, and `continuity_correction=None`.

solve_proportion ¶

solve_proportion(
    *,
    size: int,
    distance: float,
    conf_level: float = 0.95,
    interval_type: Literal[
        "two-sided", "lower one-sided", "upper one-sided"
    ] = "two-sided",
    method: Literal[
        "clopper-pearson", "wald", "wilson"
    ] = "clopper-pearson",
    continuity_correction: bool = False,
    search_direction: Literal["below", "above"] = "above",
) -> float

Estimate the required proportion, given either a desired confidence interval width or a desired distance from the proportion to the confidence bound.

Parameters:

Name	Type	Description	Default
`size`	`int`	Sample size.	required
`distance`	`float`	if `interval_type='two-sided'`, provide confidence interval width. if `interval_type='lower one-sided'` or `interval_type='upper one-sided'`, provide the distance from the proportion to the confidence bound.	required
`conf_level`	`float`	Confidence level.	`0.95`
`interval_type`	`Literal['two-sided', 'lower one-sided', 'upper one-sided']`	Type of the confidence interval. `two-sided`: Two-sided confidence interval \((L, U)\). `lower one-sided`: Lower one-sided confidence interval \((L, +\infty)\). `upper one-sided`: Upper one-sided confidence interval \((-\infty, U)\).	`'two-sided'`
`method`	`Literal['clopper-pearson', 'wald', 'wilson']`	Method used in calculation of the confidence interval.	`'clopper-pearson'`
`continuity_correction`	`bool \| None`	Whether or not to apply Yate's continuity correction, only valid for `method='wald'` or `method='wilson'`.	`False`
`search_direction`	`Literal['below', 'above']`	Direction of the search. `'below'`: search for the proportion in range (0, 0.5] `'above'`: search for the proportion in range [0.5, 1)	`'above'`

Returns:

Type	Description
`float`	The required proportion.