tki.insights.shape_insight.TrendInsight

class tki.insights.shape_insight.TrendInsight(stat_distribution: ~scipy.stats._distn_infrastructure.rv_continuous = <scipy.stats._continuous_distns.logistic_gen object>, slope_mean: float = 0.2, slope_std: float = 2.0)

Trend Insights create a higher score for significant trends. The score is calculated by multiplying the impact factor with the slope and the rvalue**2 of a linear regression.

TODO: Use and compare the Results using the p-value provided by scipy.stats.linregress I would prefer using the scipy implementation

Parameters

stat_distributionscipy.stats.rv_continuous

Statistical distribution function describing the distribution of slopes. Defaults to scipy.stats.logistic

slope_meanfloat

Position of the distribution of slopes Defaults to 0.0

slope_stdfloat

Standard derivation of the distribution of slopes Defaults to 0.2

__init__(stat_distribution: ~scipy.stats._distn_infrastructure.rv_continuous = <scipy.stats._continuous_distns.logistic_gen object>, slope_mean: float = 0.2, slope_std: float = 2.0)

Methods

__init__([stat_distribution, slope_mean, ...])
calc_insight(extraction_result)	Calculate Insight score
plot(result)	Visualizes Insight Result using matplotlib

Attributes

name