interpret.slopes

interpret.slopes(
    model,
    idata,
    wrt,
    conditional=None,
    average_by=None,
    eps=0.0001,
    slope='dydx',
    use_hdi=True,
    prob=None,
    transforms=None,
    sample_new_groups=False,
)

Compute Conditional Adjusted Slopes

Parameters

model : bambi.Model

The model for which we want to plot the predictions.

idata : arviz.InferenceData

The InferenceData object that contains the samples from the posterior distribution of the model.

wrt : (str, dict)

The slope of the regression with respect to (wrt) this predictor will be computed.

conditional : (str, list, dict) = None

The covariates we would like to condition on. If dict, keys are the covariate names and values are the values to condition on.

average_by : str | list | bool | None = None

The covariates we would like to average by. The passed covariate(s) will marginalize over the other covariates in the model. If True, it averages over all covariates in the model to obtain the average estimate. Defaults to None.

eps : float = 0.0001

To compute the slope, ‘wrt’ is evaluated at wrt +/- ‘eps’. The rate of change is then computed as the difference between the two values divided by ‘eps’. Defaults to 1e-4.

slope : str = 'dydx'

The type of slope to compute. Defaults to ‘dydx’. ‘dydx’ represents a unit increase in ‘wrt’ is associated with an n-unit change in the response. ‘eyex’ represents a percentage increase in ‘wrt’ is associated with an n-percent change in the response. ‘eydx’ represents a unit increase in ‘wrt’ is associated with an n-percent change in the response. ‘dyex’ represents a percent change in ‘wrt’ is associated with a unit increase in the response.

use_hdi : bool = True

Whether to compute the highest density interval (defaults to True) or the quantiles.

prob : float = None

The probability for the credibility intervals. Must be between 0 and 1. Defaults to 0.94. Changing the global variable az.rcParams["stats.ci_prob"] affects this default.

transforms : dict = None

Transformations that are applied to each of the variables being plotted. The keys are the name of the variables, and the values are functions to be applied. Defaults to None.

sample_new_groups : bool = False

If the model contains group-level effects, and data is passed for unseen groups, whether to sample from the new groups. Defaults to False.

Returns

: pd.DataFrame

A dataframe with the comparison values, highest density interval, wrt name, contrast value, and conditional values.

Raises

: ValueError

If length of wrt is greater than 1. If conditional is None and wrt is passed more than 2 values. If conditional is None and default wrt has more than 2 unique values. If conditional is a list and the length is greater than 3. If slope is not ‘dydx’, ‘dyex’, ‘eyex’, or ‘eydx’. If prob is not > 0 and < 1.